--- title: "1.104.0rc1 - Claude Opus 5.5, GPT-6, Master Key Enforcement & Team Routing Controls" url: "/release_notes/v1.104.0rc1/v1-104-0-rc-1" canonical_url: "https://docs.litellm.ai/release_notes/v1.104.0rc1/v1-104-0-rc-1" type: "release-notes" last_updated: "2026-10-02" related: - "/release_notes/" - "/release_notes/v1.104.0rc2/v1-104-0-rc-2" --- # 1.104.0rc1 - Claude Opus 5.5, GPT-6, Master Key Enforcement & Team Routing Controls > Index of all LiteLLM docs: https://docs.litellm.ai/llms.txt ## Deploy this version **Docker** ```bash docker run \ -e LITELLM_MASTER_KEY=sk- \ -e DATABASE_URL=postgresql://:@:5432/ \ -e STORE_MODEL_IN_DB=True \ -p 4000:4000 \ docker.litellm.ai/berriai/litellm:1.104.0-rc.1 ``` **Pip** ```bash pip install litellm==1.104.0rc1 ``` The published GitHub tag is `v1.104.0-rc.1`. These notes compare it with `v1.103.0-rc.1`, the previous release candidate cut from `main`. Changes backported onto `rc/1.103.0` and already shipped in `v1.103.0` are omitted, and so are the three Usage page follow-ups built on the top-N key cap, which were reverted on `rc/1.104.0` before this tag so the Usage pages load every key the same way `v1.103.0` does Customer-facing changes come first. Test, CI and internal changes are listed at the bottom :::danger[Breaking Changes] These callouts cover user-facing behavior that differs from `v1.103.0`, the latest stable release **The proxy refuses to start with an unset, empty, or publicly known master key.** A deployment with no `LITELLM_MASTER_KEY`, an empty one, or `sk-1234` stops booting after the upgrade. The startup error names where the bad key came from and prints a command that generates a secure one. If the database holds values encrypted with the old key, also set `LITELLM_MIGRATE_FROM_MASTER_KEY` so the next boot re-encrypts them. To keep the old behavior on a local sandbox, set `LITELLM_DANGEROUSLY_PERMIT_WEAK_OR_UNSET_MASTER_KEY=true` or `general_settings.dangerously_permit_weak_or_unset_master_key: true`. See [PR #42019](https://github.com/BerriAI/litellm/pull/42019), [PR #42011](https://github.com/BerriAI/litellm/pull/42011) **An exhausted budget now returns HTTP 422 instead of 429.** Clients stop treating a spent budget as a retryable rate limit. Real rpm/tpm limits still return 429. Set `litellm_settings.budget_exceeded_status_code: 429` to keep the old status. See [PR #42097](https://github.com/BerriAI/litellm/pull/42097) ::: ## Key Highlights - **New frontier models on day one**: Claude Opus 5.5 across Anthropic, Bedrock, Vertex AI and Azure AI, and GPT-6 Sol and GPT-6 Luna across OpenAI, Bedrock and Azure Foundry, among 331 new catalog entries - **New providers and routes**: Eden AI and Nadir providers, TinyFish, fal.ai queue and OpenRouter decisions pass-through routes, OpenAI models on Bedrock's native Responses API, and Claude on Bedrock Mantle's native Messages API - **Gateway hardening**: the proxy refuses a weak or missing master key, dashboard sign-in adds breached-password detection and forced password resets, sessions are revoked on logout, and auth fails closed during a database outage - **Routing controls**: group-scoped priority routing, time-windowed team reservation of deployments, native compact-to-fit across conversation APIs, a JEV classifier for the Auto Router, and configurable provider affinity headers - **Admin UI**: the LiteAdmin assistant, prompt caching savings, internal-user savings and Auto Router usage, and Capability and Fuse v2 routing forecasts ## New Providers and Endpoints ### New Providers (2 new providers) | Provider | Supported LiteLLM Endpoints | Description | | --- | --- | --- | | [Eden AI](../../docs/providers/edenai) | `/v1/chat/completions`, `/v1/responses`, `/v1/messages`, `/v1/embeddings`, audio, images, `/v1/videos` | Eden AI's unified API across chat, embeddings, audio, image and video models | | [Nadir](../../docs/providers/nadir) | `/v1/chat/completions` | Nadir's `nadir/auto` router, with the cost Nadir reports recorded as the response cost | ### Expanded provider endpoint support | Provider | Endpoint | What you can do | | --- | --- | --- | | [TinyFish](../../docs/pass_through/tinyfish) | `/tinyfish/*` | Run TinyFish Agent API automations through a pass-through route with per-step billing | | [fal.ai](../../docs/providers/fal_ai) | `/fal_ai/*` | Submit and poll fal queue jobs through a pass-through route with spend tracking | | [OpenRouter](../../docs/providers/openrouter) | `/openrouter/alpha/decisions` | Reach OpenRouter's decisions API through a pass-through route | | [Amazon Bedrock](../../docs/providers/bedrock) | `/v1/responses` | Serve OpenAI models on bedrock-runtime's native Responses API | | [Bedrock Mantle](../../docs/providers/bedrock_mantle) | `/v1/messages` | Serve Claude models on Mantle's native Anthropic Messages API | | [Vertex AI](../../docs/providers/vertex_batch) | `/v1/files`, `/v1/batches` | Submit native Vertex batch JSONL with cost tracking | ## New Models / Updated Models #### New Model Support (331 new models) Counts represent new catalog identifiers, including aliases and regional variants. Prices below are the values bundled in this release, in USD; runtime pricing-map reloads can update them. Input and output columns show base token rates; long-context, cache, image-token, and other specialized rates depend on the model | Provider | Model | Context Window | Input ($/1M tokens) | Output ($/1M tokens) | Features / special pricing | | --- | --- | --- | --- | --- | --- | | Amazon Bedrock | `anthropic.claude-mythos-5-1` | 1,000,000 | $10 | $50 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `anthropic.claude-opus-5-5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-fable-5` | 1,000,000 | $11 | $55 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-mythos-preview` | 1,000,000 | $27.5 | $137.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `apac.anthropic.claude-opus-4-7` | 1,000,000 | $5.5 | $27.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-opus-4-8` | 1,000,000 | $5.5 | $27.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-opus-5` | 1,000,000 | $5.5 | $27.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-opus-5-5` | 1,000,000 | $4.4 | $22 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-sonnet-4-6` | 1,000,000 | $3.3 | $16.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `apac.anthropic.claude-sonnet-5` | 1,000,000 | $2.2 | $11 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `au.anthropic.claude-fable-5` | 1,000,000 | $11 | $55 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `au.anthropic.claude-mythos-preview` | 1,000,000 | $27.5 | $137.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `au.anthropic.claude-opus-5-5` | 1,000,000 | $4.4 | $22 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `bedrock/ap-northeast-1/qwen.qwen3-next-80b-a3b` | 128,000 | $0.18 | $1.45 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/ap-south-1/qwen.qwen3-next-80b-a3b` | 128,000 | $0.18 | $1.41 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/ap-southeast-2/qwen.qwen3-next-80b-a3b` | 128,000 | $0.1545 | $1.236 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/eu-west-1/qwen.qwen3-next-80b-a3b` | 128,000 | $0.18 | $1.41 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/eu-west-2/nvidia.nemotron-super-3-120b` | 256,000 | $0.23 | $1.01 | Chat; Reasoning; Tool calling; Structured output | | Amazon Bedrock | `bedrock/eu-west-2/qwen.qwen3-next-80b-a3b` | 128,000 | $0.23 | $1.86 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/sa-east-1/qwen.qwen3-next-80b-a3b` | 128,000 | $0.18 | $1.45 | Chat; Tool calling; Structured output | | Amazon Bedrock | `bedrock/us-gov-east-1/anthropic.claude-opus-5-5` | 1,000,000 | $4.8 | $24 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `bedrock/us-gov-west-1/anthropic.claude-opus-5-5` | 1,000,000 | $4.8 | $24 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `deepseek.r1-v1:0` | 128,000 | $1.35 | $5.4 | Chat; Reasoning | | Amazon Bedrock | `eu.anthropic.claude-opus-5-5` | 1,000,000 | $4.4 | $22 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `global.anthropic.claude-mythos-5-1` | 1,000,000 | $10 | $50 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `global.anthropic.claude-opus-5-5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `global.moonshotai.kimi-k3` | 1,000,000 | $3 | $15 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Amazon Bedrock | `global.openai.gpt-5.4` | 1,000,000 | $2.5 | $15 | Chat; Reasoning; Vision; Tool calling | | Amazon Bedrock | `global.openai.gpt-5.5` | 1,000,000 | $5 | $30 | Chat; Reasoning; Vision; Tool calling | | Amazon Bedrock | `global.openai.gpt-6-luna` | 1,050,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `global.openai.gpt-6-sol` | 1,050,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `jp.anthropic.claude-opus-5-5` | 1,000,000 | $4.4 | $22 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `mistral.pixtral-large-2502-v1:0` | 128,000 | $2 | $6 | Chat; Tool calling | | Amazon Bedrock | `moonshotai.kimi-k3` | 1,000,000 | $3 | $15 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Amazon Bedrock | `openai.gpt-6-luna` | 1,050,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `openai.gpt-6-sol` | 1,050,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `us-gov.anthropic.claude-opus-5-5` | 1,000,000 | $4.8 | $24 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `us.anthropic.claude-mythos-5` | 1,000,000 | $11 | $55 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `us.anthropic.claude-mythos-5-1` | 1,000,000 | $11 | $55 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `us.anthropic.claude-mythos-preview` | 1,000,000 | $27.5 | $137.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `us.anthropic.claude-opus-5-5` | 1,000,000 | $4.4 | $22 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Amazon Bedrock | `us.moonshotai.kimi-k3` | 1,000,000 | $3.3 | $16.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Amazon Bedrock | `us.openai.gpt-5.4` | 1,000,000 | $2.75 | $16.5 | Chat; Reasoning; Vision; Tool calling | | Amazon Bedrock | `us.openai.gpt-5.5` | 1,000,000 | $5.5 | $33 | Chat; Reasoning; Vision; Tool calling | | Amazon Bedrock | `us.openai.gpt-6-luna` | 1,050,000 | $0.11 | $0.55 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Amazon Bedrock | `us.openai.gpt-6-sol` | 1,050,000 | $2.2 | $11 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Anthropic | `claude-opus-5-5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Azure AI | `azure_ai/Cohere-command-a-plus-05-2026` | 128,000 | $0.8 | $3.2 | Chat; Reasoning; Tool calling | | Azure AI | `azure_ai/FW-DeepSeek-V4-Flash` | 1,000,000 | $0.15 | $0.31 | Chat; Reasoning; Tool calling; Prompt caching | | Azure AI | `azure_ai/FW-DeepSeek-V4.1-Flash` | 1,000,000 | $0.375 | $1.5 | Chat; Reasoning; Tool calling; Prompt caching | | Azure AI | `azure_ai/FW-GLM-5.3` | 1,048,576 | $1.75 | $5.5 | Chat; Reasoning; Tool calling; Prompt caching | | Azure AI | `azure_ai/FW-GLM-5.3-Flash` | 1,048,576 | $0.188 | $0.625 | Chat; Reasoning; Tool calling; Prompt caching | | Azure AI | `azure_ai/FW-GPT-OSS-120B` | 131,072 | $0.165 | $0.66 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Azure AI | `azure_ai/MAI-Image-2.5-Pro` | - | $5 | - | Image generation | | Azure AI | `azure_ai/MAI-Image-2.6` | - | $5 | - | Image generation | | Azure AI | `azure_ai/MAI-Image-2.6-Flash` | - | $1.75 | - | Image generation | | Azure AI | `azure_ai/claude-opus-5-5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Azure AI | `azure_ai/deepseek-v4.1-flash` | 1,000,000 | $0.375 | $1.5 | Chat; Reasoning; Tool calling; Prompt caching | | Azure AI | `azure_ai/gpt-6-luna` | 922,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure AI | `azure_ai/gpt-6-sol` | 922,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure AI | `azure_ai/gpt-image-2` | - | $5 | - | Image generation; Vision | | Azure AI | `azure_ai/mistral-medium-3-5` | 128,000 | $1.5 | $7.5 | Chat; Vision; Structured output | | Azure AI | `azure_ai/muse-spark-1.3` | 1,048,576 | $1.25 | $4.25 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Azure OpenAI | `azure/eu/gpt-6-luna` | 922,000 | $0.12 | $0.6 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/eu/gpt-6-sol` | 922,000 | $2.4 | $12 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/gpt-6-luna` | 922,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/gpt-6-luna-2026-09-22` | 922,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/gpt-6-sol` | 922,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/gpt-6-sol-2026-09-22` | 922,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/gpt-audio` | 128,000 | $2.5 | $10 | Chat; Tool calling | | Azure OpenAI | `azure/gpt-audio-1.5` | 128,000 | $2.5 | $10 | Chat; Tool calling | | Azure OpenAI | `azure/gpt-live-1` | - | - | - | Realtime; Tool calling; Audio input; `input_cost_per_second`: $0.00083333 | | Azure OpenAI | `azure/gpt-live-transcribe` | 32,000 | - | - | Transcription; Audio input; `input_cost_per_second`: $0.00028333 | | Azure OpenAI | `azure/gpt-realtime` | 32,000 | $4 | $16 | Realtime; Tool calling; Audio input | | Azure OpenAI | `azure/gpt-realtime-1.5` | 32,000 | $4 | $16 | Realtime; Tool calling; Audio input | | Azure OpenAI | `azure/gpt-realtime-translate` | 32,000 | - | - | Realtime; Audio input; `input_cost_per_second`: $0.00056667 | | Azure OpenAI | `azure/gpt-transcribe` | - | - | - | Transcription; Audio input; `input_cost_per_second`: $0.000075 | | Azure OpenAI | `azure/us/gpt-6-luna` | 922,000 | $0.11 | $0.55 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Azure OpenAI | `azure/us/gpt-6-sol` | 922,000 | $2.2 | $11 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | Baseten | `baseten/deepseek-ai/DeepSeek-V4-Flash-0731` | 1,048,576 | $0.13 | $0.26 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/deepseek-ai/DeepSeek-V4-Pro` | 1,048,576 | $1.74 | $3.48 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/deepseek-ai/DeepSeek-V4-Pro-0813` | 1,048,576 | $1.32 | $3.96 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/deepseek-ai/DeepSeek-V4.1-Flash` | 1,048,576 | $0.3 | $1.2 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/moonshotai/Kimi-K2.6` | 262,000 | $0.95 | $4 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/moonshotai/Kimi-K2.7-Code` | 262,000 | $0.95 | $4 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/moonshotai/Kimi-K3` | 1,048,576 | $3 | $15 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/nvidia/NVIDIA-Nemotron-3-Ultra-550B-A55B` | 202,800 | $0.6 | $2.4 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/thinkingmachines/inkling` | 1,048,576 | $1 | $4.05 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/thinkingmachines/inkling-small` | 1,048,576 | $0.5 | $1.2 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/zai-org/GLM-5.2` | 1,048,576 | $1.4 | $4.4 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/zai-org/GLM-5.2-Fast` | 1,048,576 | $2.1 | $6.6 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/zai-org/GLM-5.3-Fast` | 1,048,576 | $2.1 | $6.6 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Baseten | `baseten/zai-org/GLM-5.3-Flash` | 1,048,576 | $0.15 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Bedrock Mantle | `bedrock_mantle/anthropic.claude-haiku-4-5` | 200,000 | $1 | $5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Bedrock Mantle | `bedrock_mantle/deepseek.v3.1` | 128,000 | $0.58 | $1.68 | Chat; Tool calling | | Bedrock Mantle | `bedrock_mantle/moonshotai.kimi-k2-thinking` | 256,000 | $0.6 | $2.5 | Chat; Reasoning; Tool calling | | Bedrock Mantle | `bedrock_mantle/openai.gpt-6-luna` | 1,050,000 | $0.11 | $0.55 | Responses; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Bedrock Mantle | `bedrock_mantle/openai.gpt-6-sol` | 1,050,000 | $2.2 | $11 | Responses; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-235b-a22b-2507` | 256,000 | $0.22 | $0.88 | Chat; Reasoning; Tool calling | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-32b` | 32,000 | $0.15 | $0.6 | Chat; Reasoning; Tool calling | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-coder-30b-a3b-instruct` | 256,000 | $0.15 | $0.6 | Chat; Tool calling | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-coder-480b-a35b-instruct` | 128,000 | $0.45 | $1.8 | Chat; Tool calling | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-next-80b-a3b-instruct` | 256,000 | $0.14 | $1.2 | Chat; Reasoning; Tool calling | | Bedrock Mantle | `bedrock_mantle/qwen.qwen3-vl-235b-a22b-instruct` | 256,000 | $0.53 | $2.66 | Chat; Vision; Tool calling | | Cohere | `c4ai-aya-expanse-32b` | 128,000 | $0.5 | $1.5 | Chat | | Fireworks AI | `fireworks_ai/accounts/fireworks/models/ember-1` | 1,048,576 | $3 | $15 | Chat; Reasoning; Vision; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/accounts/fireworks/routers/deepseek-v4p1-flash-us` | 1,048,576 | $0.45 | $1.8 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Fireworks AI | `fireworks_ai/accounts/fireworks/routers/glm-5p3-flash-us` | 1,048,576 | $0.225 | $0.75 | Chat; Vision; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/accounts/fireworks/routers/glm-5p3-us` | 1,048,576 | $2.1 | $6.6 | Chat; Reasoning; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/deepseek-v4-pro-0813` | 1,048,576 | $1.32 | $3.96 | Chat; Reasoning; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/deepseek-v4p1-flash-us` | 1,048,576 | $0.45 | $1.8 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | Fireworks AI | `fireworks_ai/glm-5p3` | 1,048,576 | $1.4 | $4.4 | Chat; Reasoning; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/glm-5p3-flash` | 1,048,576 | $0.15 | $0.5 | Chat; Vision; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/glm-5p3-flash-us` | 1,048,576 | $0.225 | $0.75 | Chat; Vision; Tool calling; Structured output | | Fireworks AI | `fireworks_ai/glm-5p3-us` | 1,048,576 | $2.1 | $6.6 | Chat; Reasoning; Tool calling; Structured output | | Gemini | `gemini/deep-research-max-preview-04-2026` | 131,072 | $2 | $12 | Chat; Vision; Web search | | Gemini | `gemini/deep-research-preview-04-2026` | 131,072 | $2 | $12 | Chat; Vision; Web search | | Gemini | `gemini/gemini-3.8-flash-lite-tts` | 8,192 | $0.5 | $6 | Speech | | Gemini | `gemini/gemini-3.8-flash-tts` | 8,192 | $0.5 | $9 | Speech | | Gemini | `gemini/lyria-realtime-exp` | 1,048,576 | $0 | $0 | Chat | | Groq | `groq/llama-guard-3-8b` | 8,192 | $0.2 | $0.2 | Chat | | Nebius | `nebius/deepseek-ai/DeepSeek-V4.1-Flash` | 1,048,576 | $0.3 | $1.2 | Chat; Vision | | OpenAI | `gpt-6-luna` | 922,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | OpenAI | `gpt-6-sol` | 922,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search; Computer use | | OpenRouter | `openrouter/aion-labs/aion-3.5` | 262,144 | $3 | $6 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/aion-labs/aion-3.5-mini` | 262,144 | $0.7 | $1.4 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/anthropic/claude-opus-5.5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/anthropic/claude-opus-5.5:batch` | 1,000,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/cohere/command-a-plus` | 192,000 | $0.3 | $1.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/deepseek/deepseek-v4.1-flash:batch` | 1,048,576 | $0.112 | $0.336 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/fireworks/ember-1` | 1,048,576 | $3 | $15 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/nex-agi/nex-n2.5-mini` | 262,144 | $0.025 | $0.1 | Chat; Reasoning; Vision; Prompt caching; Structured output | | OpenRouter | `openrouter/nex-agi/nex-n2.5-pro` | 262,144 | $0.075 | $0.25 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/openai/gpt-6-luna` | 1,050,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-luna-pro` | 1,050,000 | $0.1 | $0.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-luna-pro:batch` | 1,050,000 | $0.05 | $0.25 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-luna:batch` | 1,050,000 | $0.05 | $0.25 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-sol` | 1,050,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-sol-pro` | 1,050,000 | $2 | $10 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-sol-pro:batch` | 1,050,000 | $1 | $5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-6-sol:batch` | 1,050,000 | $1 | $5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/openai/gpt-oss-20b:batch` | 131,072 | $0.024 | $0.112 | Chat; Reasoning; Tool calling; Structured output | | OpenRouter | `openrouter/perceptron/perceptron-mk1.5` | 36,864 | $0.15 | $1.5 | Chat; Reasoning; Vision; Tool calling; Structured output; Audio input; Video input | | OpenRouter | `openrouter/qwen/qwen3.8-max-prime` | 1,000,000 | $4 | $12 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Video input | | OpenRouter | `openrouter/qwen/qwen3.8-omni-flash` | 1,000,000 | $0.15 | $0.47 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input | | OpenRouter | `openrouter/stealth/space-bunny-alpha` | 1,000,000 | $0 | $0 | Chat; Reasoning; Vision; Tool calling; Video input | | OpenRouter | `openrouter/typesafe/jev-1.13` | 32,000 | $0.042 | $0 | evaluation | | OpenRouter | `openrouter/upstage/solar-mini4` | 524,288 | $0.05 | $0.2 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | OpenRouter | `openrouter/x-ai/grok-4.7` | 500,000 | $1.6 | $4.8 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Web search | | OpenRouter | `openrouter/xiaomi/mimo-v2.6-flash` | 1,048,576 | $0.14 | $0.28 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input | | OpenRouter | `openrouter/xiaomi/mimo-v2.6-pro` | 1,048,576 | $0.435 | $0.87 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input | | OpenRouter | `openrouter/xiaomi/mimo-v2.6-pro-ultraspeed` | 1,048,576 | $4.35 | $8.7 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input | | OpenRouter | `openrouter/z-ai/glm-5.3-prime` | 1,000,000 | $2.8 | $8.8 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Together AI | `together_ai/together/Tev1-4B-experimental` | 32,768 | $0.042 | $0 | Chat | | Vertex AI | `gemini-3.8-flash-cyber` | 1,048,576 | $1.5 | $7.5 | Chat; Reasoning; Vision; Prompt caching; Structured output; PDF input; Audio input; Video input | | Vertex AI | `vertex_ai/chirp_2` | - | - | - | Transcription; `input_cost_per_second`: $0.00026667 | | Vertex AI | `vertex_ai/claude-opus-5-5` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Vertex AI | `vertex_ai/claude-opus-5-5@default` | 1,000,000 | $4 | $20 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; PDF input; Computer use | | Vertex AI | `vertex_ai/gemini-2.0-flash` | - | $0.15 | $0.6 | Chat | | Vertex AI | `vertex_ai/gemini-2.0-flash-lite` | - | $0.075 | $0.3 | Chat | | Vertex AI | `vertex_ai/gemini-2.5-flash-tts` | - | $0.5 | $10 | Speech | | Vertex AI | `vertex_ai/gemini-2.5-pro-tts` | - | $1 | $20 | Speech | | Vertex AI | `vertex_ai/gemini-3.8-flash-cyber` | 1,048,576 | $1.5 | $7.5 | Chat; Reasoning; Vision; Prompt caching; Structured output; PDF input; Audio input; Video input | | Vertex AI | `vertex_ai/gemini-3.8-live` | 131,072 | $0.75 | $4.5 | Realtime; Vision; Tool calling; Web search; Audio input | | Vertex AI | `vertex_ai/gemini-omni-1.1-flash-preview` | 57,920 | $1.5 | $9 | Chat; Reasoning; Vision; Video input | | Vertex AI | `vertex_ai/meta/llama-3.3-70b-instruct-maas` | 128,000 | $0.72 | $0.72 | Chat; Tool calling | | Vertex AI | `vertex_ai/virtual-try-on-001` | - | - | - | Image generation; `output_cost_per_image`: $0.06 | | Vertex AI | `vertex_ai/zai-org/glm-5.2-maas` | 1,000,000 | $1.4 | $4.4 | Chat; Reasoning; Tool calling; Prompt caching; Structured output | | Weights & Biases | `wandb/deepseek-ai/DeepSeek-V4.1-Flash` | 1,049,000 | $0.2 | $0.65 | Chat; Reasoning; Vision; Prompt caching | | Weights & Biases | `wandb/google/gemma-4-26B-A4B-it` | 262,000 | $0.1 | $0.3 | Chat; Reasoning; Vision; Tool calling; Prompt caching | | Xiaomi MiMo | `xiaomi_mimo/mimo-v2.6-flash` | 1,048,576 | $0.14 | $0.28 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input; Video input | | Xiaomi MiMo | `xiaomi_mimo/mimo-v2.6-pro` | 1,048,576 | $0.435 | $0.87 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Audio input; Video input | | fal.ai | `fal_ai/bytedance/seedance-2.0/image-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.3034; `output_cost_per_second_480p`: $0.1346; `output_cost_per_second_720p`: $0.3034 | | fal.ai | `fal_ai/bytedance/seedance-2.0/reference-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.3034; `output_cost_per_second_480p`: $0.1346; `output_cost_per_second_720p`: $0.3034 | | fal.ai | `fal_ai/bytedance/seedance-2.0/text-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.3034; `output_cost_per_second_480p`: $0.1346; `output_cost_per_second_720p`: $0.3034 | | fal.ai | `fal_ai/bytedance/seedance-2.5/image-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.473; `output_cost_per_second_480p`: $0.2205; `output_cost_per_second_720p`: $0.473 | | fal.ai | `fal_ai/bytedance/seedance-2.5/reference-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.473; `output_cost_per_second_480p`: $0.2205; `output_cost_per_second_720p`: $0.473 | | fal.ai | `fal_ai/bytedance/seedance-2.5/text-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.473; `output_cost_per_second_480p`: $0.2205; `output_cost_per_second_720p`: $0.473 | | fal.ai | `fal_ai/fal-ai/flux-lora-depth` | - | - | - | Image generation; `output_cost_per_image`: $0.035; `output_cost_per_pixel`: $0.00000003 | | fal.ai | `fal_ai/fal-ai/flux/dev` | - | - | - | Image generation; `output_cost_per_image`: $0.025; `output_cost_per_pixel`: $0.00000002 | | fal.ai | `fal_ai/fal-ai/moondream3-preview/query` | - | $0.4 | $3.5 | Chat; Reasoning; Vision | | fal.ai | `fal_ai/fal-ai/nano-banana-2` | - | - | - | Image generation; `output_cost_per_image`: $0.08; `output_cost_per_image_0.5K`: $0.06; `output_cost_per_image_1K`: $0.08 | | fal.ai | `fal_ai/fal-ai/nano-banana-pro` | - | - | - | Image generation; `output_cost_per_image`: $0.15; `output_cost_per_image_1K`: $0.15; `output_cost_per_image_2K`: $0.15 | | fal.ai | `fal_ai/fal-ai/trellis` | - | - | - | Image generation; `output_cost_per_image`: $0.02 | | fal.ai | `fal_ai/fal-ai/trellis-2` | - | - | - | Image generation; `output_cost_per_image`: $0.3; `output_cost_per_image_512`: $0.25; `output_cost_per_image_1024`: $0.3 | | fal.ai | `fal_ai/high/1024-x-1024/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05268 | | fal.ai | `fal_ai/high/1024-x-1024/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05268 | | fal.ai | `fal_ai/high/1024-x-1024/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05268 | | fal.ai | `fal_ai/high/1024-x-1024/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05268 | | fal.ai | `fal_ai/high/1024-x-1536/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.04116 | | fal.ai | `fal_ai/high/1024-x-1536/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.04116 | | fal.ai | `fal_ai/high/1024-x-1536/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.04116 | | fal.ai | `fal_ai/high/1024-x-1536/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.04116 | | fal.ai | `fal_ai/high/1024-x-768/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/high/1024-x-768/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/high/1024-x-768/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/high/1024-x-768/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/high/1920-x-1080/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0396 | | fal.ai | `fal_ai/high/1920-x-1080/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0396 | | fal.ai | `fal_ai/high/1920-x-1080/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0396 | | fal.ai | `fal_ai/high/1920-x-1080/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0396 | | fal.ai | `fal_ai/high/2560-x-1440/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05529 | | fal.ai | `fal_ai/high/2560-x-1440/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05529 | | fal.ai | `fal_ai/high/2560-x-1440/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05529 | | fal.ai | `fal_ai/high/2560-x-1440/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.05529 | | fal.ai | `fal_ai/high/3840-x-2160/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.10008 | | fal.ai | `fal_ai/high/3840-x-2160/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.10008 | | fal.ai | `fal_ai/high/3840-x-2160/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.10008 | | fal.ai | `fal_ai/high/3840-x-2160/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.10008 | | fal.ai | `fal_ai/low/1024-x-1024/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00588 | | fal.ai | `fal_ai/low/1024-x-1024/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00588 | | fal.ai | `fal_ai/low/1024-x-1024/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00588 | | fal.ai | `fal_ai/low/1024-x-1024/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00588 | | fal.ai | `fal_ai/low/1024-x-1536/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00474 | | fal.ai | `fal_ai/low/1024-x-1536/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00474 | | fal.ai | `fal_ai/low/1024-x-1536/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00474 | | fal.ai | `fal_ai/low/1024-x-1536/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00474 | | fal.ai | `fal_ai/low/1024-x-768/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00402 | | fal.ai | `fal_ai/low/1024-x-768/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00402 | | fal.ai | `fal_ai/low/1024-x-768/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00402 | | fal.ai | `fal_ai/low/1024-x-768/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00402 | | fal.ai | `fal_ai/low/1920-x-1080/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00441 | | fal.ai | `fal_ai/low/1920-x-1080/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00441 | | fal.ai | `fal_ai/low/1920-x-1080/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00441 | | fal.ai | `fal_ai/low/1920-x-1080/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00441 | | fal.ai | `fal_ai/low/2560-x-1440/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00615 | | fal.ai | `fal_ai/low/2560-x-1440/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00615 | | fal.ai | `fal_ai/low/2560-x-1440/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00615 | | fal.ai | `fal_ai/low/2560-x-1440/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00615 | | fal.ai | `fal_ai/low/3840-x-2160/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01113 | | fal.ai | `fal_ai/low/3840-x-2160/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01113 | | fal.ai | `fal_ai/low/3840-x-2160/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01113 | | fal.ai | `fal_ai/low/3840-x-2160/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01113 | | fal.ai | `fal_ai/max/1024-x-1024/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.21072 | | fal.ai | `fal_ai/max/1024-x-1024/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.21072 | | fal.ai | `fal_ai/max/1024-x-1024/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.21072 | | fal.ai | `fal_ai/max/1024-x-1024/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.21072 | | fal.ai | `fal_ai/max/1024-x-1536/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.16464 | | fal.ai | `fal_ai/max/1024-x-1536/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.16464 | | fal.ai | `fal_ai/max/1024-x-1536/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.16464 | | fal.ai | `fal_ai/max/1024-x-1536/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.16464 | | fal.ai | `fal_ai/max/1024-x-768/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.14445 | | fal.ai | `fal_ai/max/1024-x-768/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.14445 | | fal.ai | `fal_ai/max/1024-x-768/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.14445 | | fal.ai | `fal_ai/max/1024-x-768/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.14445 | | fal.ai | `fal_ai/max/1920-x-1080/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1584 | | fal.ai | `fal_ai/max/1920-x-1080/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1584 | | fal.ai | `fal_ai/max/1920-x-1080/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1584 | | fal.ai | `fal_ai/max/1920-x-1080/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1584 | | fal.ai | `fal_ai/max/2560-x-1440/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.2211 | | fal.ai | `fal_ai/max/2560-x-1440/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.2211 | | fal.ai | `fal_ai/max/2560-x-1440/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.2211 | | fal.ai | `fal_ai/max/2560-x-1440/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.2211 | | fal.ai | `fal_ai/max/3840-x-2160/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.40026 | | fal.ai | `fal_ai/max/3840-x-2160/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.40026 | | fal.ai | `fal_ai/max/3840-x-2160/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.40026 | | fal.ai | `fal_ai/max/3840-x-2160/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.40026 | | fal.ai | `fal_ai/medium/1024-x-1024/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01317 | | fal.ai | `fal_ai/medium/1024-x-1024/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01317 | | fal.ai | `fal_ai/medium/1024-x-1024/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01317 | | fal.ai | `fal_ai/medium/1024-x-1024/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01317 | | fal.ai | `fal_ai/medium/1024-x-1536/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1024-x-1536/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1024-x-1536/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1024-x-1536/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1024-x-768/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00903 | | fal.ai | `fal_ai/medium/1024-x-768/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00903 | | fal.ai | `fal_ai/medium/1024-x-768/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00903 | | fal.ai | `fal_ai/medium/1024-x-768/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.00903 | | fal.ai | `fal_ai/medium/1920-x-1080/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1920-x-1080/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1920-x-1080/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/1920-x-1080/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01029 | | fal.ai | `fal_ai/medium/2560-x-1440/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01434 | | fal.ai | `fal_ai/medium/2560-x-1440/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01434 | | fal.ai | `fal_ai/medium/2560-x-1440/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01434 | | fal.ai | `fal_ai/medium/2560-x-1440/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.01434 | | fal.ai | `fal_ai/medium/3840-x-2160/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.02595 | | fal.ai | `fal_ai/medium/3840-x-2160/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.02595 | | fal.ai | `fal_ai/medium/3840-x-2160/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.02595 | | fal.ai | `fal_ai/medium/3840-x-2160/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.02595 | | fal.ai | `fal_ai/minimax/h3/reference-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.13; `output_cost_per_second_480p`: $0.05; `output_cost_per_second_768p`: $0.06 | | fal.ai | `fal_ai/minimax/h3/text-to-video` | - | - | - | Video generation; `output_cost_per_second`: $0.13; `output_cost_per_second_480p`: $0.05; `output_cost_per_second_768p`: $0.06 | | fal.ai | `fal_ai/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.03612 | | fal.ai | `fal_ai/xhigh/1024-x-1024/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09366 | | fal.ai | `fal_ai/xhigh/1024-x-1024/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09366 | | fal.ai | `fal_ai/xhigh/1024-x-1024/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09366 | | fal.ai | `fal_ai/xhigh/1024-x-1024/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09366 | | fal.ai | `fal_ai/xhigh/1024-x-1536/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07377 | | fal.ai | `fal_ai/xhigh/1024-x-1536/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07377 | | fal.ai | `fal_ai/xhigh/1024-x-1536/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07377 | | fal.ai | `fal_ai/xhigh/1024-x-1536/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07377 | | fal.ai | `fal_ai/xhigh/1024-x-768/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0642 | | fal.ai | `fal_ai/xhigh/1024-x-768/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0642 | | fal.ai | `fal_ai/xhigh/1024-x-768/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0642 | | fal.ai | `fal_ai/xhigh/1024-x-768/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.0642 | | fal.ai | `fal_ai/xhigh/1920-x-1080/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07041 | | fal.ai | `fal_ai/xhigh/1920-x-1080/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07041 | | fal.ai | `fal_ai/xhigh/1920-x-1080/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07041 | | fal.ai | `fal_ai/xhigh/1920-x-1080/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.07041 | | fal.ai | `fal_ai/xhigh/2560-x-1440/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09828 | | fal.ai | `fal_ai/xhigh/2560-x-1440/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09828 | | fal.ai | `fal_ai/xhigh/2560-x-1440/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09828 | | fal.ai | `fal_ai/xhigh/2560-x-1440/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.09828 | | fal.ai | `fal_ai/xhigh/3840-x-2160/openai/gpt-image-2.5/flare/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1779 | | fal.ai | `fal_ai/xhigh/3840-x-2160/openai/gpt-image-2.5/flare/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1779 | | fal.ai | `fal_ai/xhigh/3840-x-2160/openai/gpt-image-2.5/sunburst/edit` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1779 | | fal.ai | `fal_ai/xhigh/3840-x-2160/openai/gpt-image-2.5/sunburst/text-to-image` | - | - | - | Image generation; Vision; `output_cost_per_image`: $0.1779 | | xAI | `xai/grok-4.20-0309` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-0309` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-latest` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-latest-non-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-latest-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-non-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-beta-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-0304` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-0304-non-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-0304-reasoning` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-latest` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-non-reasoning-latest` | 1,000,000 | $1.25 | $2.5 | Chat; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-experimental-beta-reasoning-latest` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-multi-agent-beta-latest` | 1,000,000 | $1.25 | $2.5 | Responses; Reasoning; Vision; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-multi-agent-experimental-beta-0304` | 1,000,000 | $1.25 | $2.5 | Responses; Reasoning; Vision; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-multi-agent-experimental-beta-latest` | 1,000,000 | $1.25 | $2.5 | Responses; Reasoning; Vision; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-non-reasoning-gv2` | 1,000,000 | $1.25 | $2.5 | Chat; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.20-reasoning-gv2` | 1,000,000 | $1.25 | $2.5 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | | xAI | `xai/grok-4.7` | 500,000 | $2 | $6 | Chat; Reasoning; Vision; Tool calling; Prompt caching; Structured output; Web search | #### Updated pricing (73 models) | Provider / model | Changed token prices (USD per 1M tokens) | | --- | --- | | `anthropic.claude-mythos-preview` | Input: $0 to $27.5; Output: $0 to $137.5; Cache read: not set to $2.75; Cache write: not set to $34.375 | | `azure/eu/gpt-5.6` | Input: $5.5 to $4.4; Output: $33 to $22; Cache read: $0.55 to $0.44; Cache write: $6.875 to $5.5 | | `azure/gpt-5.6` | Input: $5 to $4; Output: $30 to $20; Cache read: $0.5 to $0.4; Cache write: $6.25 to $5 | | `azure/us/gpt-5.6` | Input: $5.5 to $4.4; Output: $33 to $22; Cache read: $0.55 to $0.44; Cache write: $6.875 to $5.5 | | `baseten/openai/gpt-oss-120b` | Cache read: not set to $0.1 | | `baseten/zai-org/GLM-4.7` | Cache read: not set to $0.12 | | `bedrock/eu-west-3/mistral.mistral-large-2402-v1:0` | Input: $10.4 to $5.2; Output: $31.2 to $15.6 | | `bedrock/us-east-1/mistral.mistral-large-2402-v1:0` | Input: $8 to $4; Output: $24 to $12 | | `bedrock/us-west-2/mistral.mistral-large-2402-v1:0` | Input: $8 to $4; Output: $24 to $12 | | `bedrock_mantle/openai.gpt-daybreak-blue-5.6-sol` | Input: $5.5 to $4.4; Output: $33 to $22; Cache read: $0.55 to $0.44; Cache write: $6.875 to $5.5 | | `cohere.command-text-v14` | Input: $1.5 to $1 | | `deepseek/deepseek-coder` | Cache read: not set to $0.014 | | `deepseek/deepseek-r1` | Cache read: not set to $0.14 | | `deepseek/deepseek-v3.2` | Cache read: not set to $0.028 | | `eu.anthropic.claude-opus-4-5-20251101-v1:0` | Input: $5 to $5.5; Output: $25 to $27.5; Cache read: $0.5 to $0.55; Cache write: $6.25 to $6.875 | | `fireworks_ai/accounts/fireworks/models/minimax-m2p7` | Input: $0.3 to not set; Output: $1.2 to not set; Cache read: $0.06 to not set | | `fireworks_ai/accounts/fireworks/routers/kimi-k3-us` | Input: $3.3 to $4.5; Output: $16.5 to $22.5; Cache read: $0.33 to $0.45 | | `fireworks_ai/kimi-k3-us` | Input: $3.3 to $4.5; Output: $16.5 to $22.5; Cache read: $0.33 to $0.45 | | `gemini-2.5-flash-image` | Cache read: $0.03 to not set | | `gemini-3-pro-image-preview` | Cache read: not set to $0.2 | | `gemini-3.1-flash-image-preview` | Cache read: not set to $0.05 | | `gemini-live-2.5-flash-preview-native-audio-09-2025` | Cache read: $0.075 to not set | | `gemini/gemini-live-2.5-flash-preview-native-audio-09-2025` | Cache read: $0.075 to not set | | `gemini/gemini-robotics-er-2-streaming-preview` | Input: $2 to $1; Output: $10 to $5 | | `mistral.mistral-large-2402-v1:0` | Input: $8 to $4; Output: $24 to $12 | | `openrouter/deepseek/deepseek-r1` | Cache read: not set to $0.14 | | `openrouter/deepseek/deepseek-v3.2-exp` | Cache read: not set to $0.02 | | `openrouter/deepseek/deepseek-v4-flash` | Input: $0.03724 to $0.049; Output: $0.07448 to $0.098; Cache read: $0.007448 to $0.0098 | | `openrouter/deepseek/deepseek-v4-flash-0731` | Input: $0.04 to $0.03; Output: $0.08 to $0.32 | | `openrouter/deepseek/deepseek-v4-flash-vision-exp` | Input: $0.2156 to $0.22; Output: $0.6468 to $0.66; Cache read: $0.00686 to $0.007 | | `openrouter/deepseek/deepseek-v4-pro` | Input: $0.422298 to $0.844944; Output: $0.844596 to $1.689888; Cache read: $0.035192 to $0.070412 | | `openrouter/deepseek/deepseek-v4-pro-0813` | Input: $0.57816 to $0.462; Output: $1.73448 to $1.386; Cache read: $0.018396 to $0.0154 | | `openrouter/meta/muse-glimmer-30b` | Input: $0.35 to $0.3; Output: $1.5 to $1.2 | | `openrouter/minimax/minimax-m2` | Input: $0.255 to $0.3; Output: $1.02 to $1.2 | | `openrouter/mistralai/mistral-large-2512` | Input: $0.55 to $0.5; Output: $1.65 to $1.5; Cache read: $0.055 to $0.05 | | `openrouter/moonshotai/kimi-k2.7-code` | Input: $0.7062 to $0.6562; Output: $3.21 to $3.3 | | `openrouter/moonshotai/kimi-k3` | Input: $1.7 to $3; Output: $8.5 to $15; Cache read: $0.17 to $0.3 | | `openrouter/moonshotai/kimi-k3:batch` | Input: $3 to $2.28; Output: $15 to $11.4; Cache read: $0.3 to $0.228 | | `openrouter/nvidia/nemotron-3-nano-30b-a3b` | Input: $0.06 to $0.05; Output: $0.24 to $0.2 | | `openrouter/nvidia/nemotron-3.5-lightning` | Input: $0.07 to $0.08 | | `openrouter/openai/gpt-oss-120b:batch` | Input: $0.15 to $0.0296; Output: $0.6 to $0.136 | | `openrouter/openai/gpt-oss-20b` | Input: $0.03 to $0.018; Output: $0.13 to $0.09 | | `openrouter/qwen/qwen3-30b-a3b-instruct-2507` | Input: $0.04815 to $0.1; Output: $0.19305 to $0.3 | | `openrouter/qwen/qwen3-next-80b-a3b-instruct` | Input: $0.09 to $0.1 | | `openrouter/qwen/qwen3.6-27b` | Input: $0.3 to $0.32; Output: $2 to $2.7; Cache read: $0.03 to $0.15 | | `openrouter/qwen/qwen3.6-35b-a3b` | Input: $0.1 to $0.15; Output: $0.9 to $1 | | `openrouter/z-ai/glm-5.2` | Input: $0.5544 to $0.6496; Output: $1.7424 to $2.0416; Cache read: $0.10296 to $0.12064 | | `openrouter/z-ai/glm-5.3` | Input: $0.896 to $1.4; Output: $2.816 to $4.4; Cache read: $0.1664 to $0.26 | | `openrouter/z-ai/glm-5.3-flash` | Input: $0.09 to $0.045; Output: $0.3 to $0.6; Cache read: $0.018 to $0.0285 | | `openrouter/z-ai/glm-5.3-flash:batch` | Input: $0.075 to $0.06; Output: $0.25 to $0.2; Cache read: $0.015 to $0.012 | | `openrouter/z-ai/glm-5.3-flashx` | Cache read: $0.075 to $0.09 | | `openrouter/z-ai/glm-5.3:batch` | Input: $0.7 to $0.45; Output: $2.2 to $2; Cache read: $0.13 to $0.1 | | `openrouter/~anthropic/claude-opus-latest` | Input: $5 to $4; Output: $25 to $20; Cache read: $0.5 to $0.2; Cache write: $6.25 to $5 | | `openrouter/~deepseek/deepseek-flash-latest` | Input: $0.13 to $0.3; Output: $0.52 to $1.2; Cache read: $0.0026 to $0.006 | | `openrouter/~deepseek/deepseek-pro-latest` | Input: $0.57816 to $1.32; Output: $1.73448 to $3.96; Cache read: $0.018396 to $0.044 | | `openrouter/~deepseek/deepseek-v4-flash-latest` | Output: $0.08 to $0.64 | | `openrouter/~moonshotai/kimi-latest` | Input: $1.7 to $3; Output: $8.5 to $15; Cache read: $0.17 to $0.3 | | `openrouter/~openai/gpt-luna-latest` | Input: $0.2 to $0.1; Output: $1.2 to $0.5; Cache read: $0.02 to $0.01; Cache write: $0.25 to $0.125 | | `openrouter/~x-ai/grok-latest` | Input: $2 to $1.6; Output: $6 to $4.8; Cache read: $0.5 to $0.4 | | `openrouter/~z-ai/glm-flash-latest` | Input: $0.075 to $0.15; Output: $0.25 to $0.5; Cache read: $0.015 to $0.05 | | `openrouter/~z-ai/glm-latest` | Input: $0.8442 to $0.6538; Output: $2.6532 to $2.0548; Cache read: $0.15678 to $0.12142 | | `together_ai/Qwen/Qwen3.7-Max` | Input: $2.5 to $1.5; Output: $7.5 to $4.5; Cache read: $0.5 to $0.3 | | `together_ai/Qwen/Qwen3.8-Flash` | Input: $0.15 to $0.09; Output: $0.47 to $0.282 | | `vertex_ai/deepseek-ai/deepseek-v3.1-maas` | Cache read: not set to $0.06 | | `vertex_ai/deepseek-ai/deepseek-v3.2-maas` | Cache read: not set to $0.056 | | `vertex_ai/gemini-2.5-flash-image` | Cache read: $0.03 to not set | | `vertex_ai/gemini-3-pro-image-preview` | Cache read: not set to $0.2 | | `vertex_ai/gemini-3.1-flash-image-preview` | Cache read: not set to $0.05 | | `vertex_ai/google/gemma-4-26b-a4b-it-maas` | Cache read: not set to $0.015 | | `vertex_ai/minimaxai/minimax-m2-maas` | Cache read: not set to $0.03 | | `vertex_ai/moonshotai/kimi-k2-thinking-maas` | Cache read: not set to $0.06 | | `vertex_ai/qwen/qwen3-coder-480b-a35b-instruct-maas` | Cache read: not set to $0.022 | | `vertex_ai/zai-org/glm-4.7-maas` | Cache read: not set to $0.06 | The registry also updates capability flags, context/output limits, non-token rates, and deprecation dates on 630 more entries
Removed catalog entries (275) `1024-x-1024/dall-e-2`, `256-x-256/dall-e-2`, `512-x-512/dall-e-2`, `amazon.nova-sonic-v1:0`, `anthropic.claude-3-haiku-20240307-v1:0`, `anthropic.claude-3-sonnet-20240229-v1:0`, `apac.anthropic.claude-3-5-sonnet-20240620-v1:0`, `apac.anthropic.claude-3-5-sonnet-20241022-v2:0`, `apac.anthropic.claude-3-haiku-20240307-v1:0`, `apac.anthropic.claude-3-sonnet-20240229-v1:0`, `azure/eu/o1-preview-2024-09-12`, `azure/global/gpt-5.1-chat`, `azure/gpt-3.5-turbo-0125`, `azure/gpt-35-turbo-0125`, `azure/gpt-35-turbo-1106`, `azure/gpt-5-chat-latest`, `azure/gpt-5.1-chat-2025-11-13`, `azure/gpt-5.2-chat-2025-12-11`, `azure/o1-preview`, `azure/o1-preview-2024-09-12`, `azure/us/o1-preview-2024-09-12`, `azure_ai/Llama-3.2-11B-Vision-Instruct`, `azure_ai/Llama-3.2-90B-Vision-Instruct`, `azure_ai/MAI-Image-2e`, `azure_ai/Meta-Llama-3.1-405B-Instruct`, `azure_ai/Meta-Llama-3.1-8B-Instruct`, `azure_ai/claude-opus-4-1`, `azure_ai/cohere-rerank-v3.5`, `azure_ai/global/grok-3`, `azure_ai/global/grok-3-mini`, `azure_ai/mistral-document-ai-2505`, `bedrock/us-gov-east-1/anthropic.claude-3-5-sonnet-20240620-v1:0`, `bedrock/us-gov-east-1/anthropic.claude-3-haiku-20240307-v1:0`, `bedrock/us-gov-west-1/anthropic.claude-3-5-sonnet-20240620-v1:0`, `bedrock/us-gov-west-1/anthropic.claude-3-7-sonnet-20250219-v1:0`, `bedrock/us-gov-west-1/anthropic.claude-3-haiku-20240307-v1:0`, `cerebras/zai-glm-4.6`, `cerebras/zai-glm-4.7`, `chatgpt-4o-latest`, `claude-3-7-sonnet-20250219`, `claude-3-haiku-20240307`, `claude-3-opus-20240229`, `claude-4-opus-20250514`, `claude-4-sonnet-20250514`, `claude-opus-4-1`, `claude-opus-4-1-20250805`, `claude-opus-4-20250514`, `claude-sonnet-4-20250514`, `codex-mini-latest`, `cohere.command-r-plus-v1:0`, `cohere.command-r-v1:0`, `command`, `command-light`, `command-r`, `command-r-plus`, `dall-e-2`, `dall-e-3`, `databricks/databricks-claude-3-7-sonnet`, `databricks/databricks-gpt-5-1-codex-max`, `databricks/databricks-gpt-5-1-codex-mini`, `databricks/databricks-gpt-5-2-codex`, `databricks/databricks-llama-2-70b-chat`, `databricks/databricks-meta-llama-3-1-405b-instruct`, `databricks/databricks-meta-llama-3-70b-instruct`, `databricks/databricks-mixtral-8x7b-instruct`, `databricks/databricks-mpt-30b-instruct`, `databricks/databricks-mpt-7b-instruct`, `deepinfra/google/gemini-2.0-flash-001`, `embed-english-light-v2.0`, `embed-english-v2.0`, `embed-multilingual-v2.0`, `eu.anthropic.claude-3-haiku-20240307-v1:0`, `eu.anthropic.claude-3-sonnet-20240229-v1:0`, `fireworks_ai/deepseek-v4-pro`, `fireworks_ai/minimax-m2p7`, `friendliai/LGAI-EXAONE/K-EXAONE-2.0-750B-A37B`, `gemini-2.0-flash`, `gemini-2.0-flash-001`, `gemini-2.0-flash-lite`, `gemini-2.0-flash-lite-001`, `gemini-2.5-flash-lite-preview-06-17`, `gemini-3-pro-preview`, `gemini/gemini-1.5-flash`, `gemini/gemini-2.0-flash`, `gemini/gemini-2.0-flash-001`, `gemini/gemini-2.0-flash-lite`, `gemini/gemini-2.0-flash-lite-001`, `gemini/gemini-2.5-flash-lite-preview-06-17`, `gemini/gemini-2.5-flash-lite-preview-09-2025`, `gemini/gemini-2.5-flash-preview-09-2025`, `gemini/gemini-3-pro-preview`, `gemini/gemini-robotics-er-1.5-preview`, `gemini/gemini-robotics-er-1.6-preview`, `gemini/imagen-3.0-generate-002`, `gemini/imagen-4.0-fast-generate-001`, `gemini/imagen-4.0-generate-001`, `gemini/imagen-4.0-ultra-generate-001`, `gemini/veo-2.0-generate-001`, `gpt-4-0125-preview`, `gpt-4-0314`, `gpt-4-turbo-preview`, `gpt-4o-audio-preview`, `gpt-4o-mini-audio-preview`, `gpt-4o-mini-realtime-preview`, `gpt-4o-mini-search-preview-2025-03-11`, `gpt-4o-mini-tts-2025-03-20`, `gpt-4o-realtime-preview`, `gpt-4o-realtime-preview-2024-12-17`, `gpt-4o-realtime-preview-2025-06-03`, `gpt-4o-search-preview-2025-03-11`, `gpt-audio-mini-2025-10-06`, `gpt-realtime-mini-2025-10-06`, `groq/gemma-7b-it`, `groq/llama-3.1-8b-instant`, `groq/llama-3.3-70b-versatile`, `groq/meta-llama/llama-4-maverick-17b-128e-instruct`, `groq/meta-llama/llama-4-scout-17b-16e-instruct`, `groq/meta-llama/llama-guard-4-12b`, `groq/moonshotai/kimi-k2-instruct-0905`, `groq/playai-tts`, `groq/qwen/qwen3-32b`, `groq/qwen/qwen3.6-27b`, `hd/1024-x-1024/dall-e-3`, `hd/1024-x-1792/dall-e-3`, `hd/1792-x-1024/dall-e-3`, `mistral/codestral-2405`, `mistral/devstral-2512`, `mistral/devstral-medium-2507`, `mistral/devstral-small-2505`, `mistral/devstral-small-2507`, `mistral/labs-devstral-small-2512`, `mistral/magistral-medium-1-2-2509`, `mistral/magistral-medium-2506`, `mistral/magistral-medium-2509`, `mistral/magistral-small-1-2-2509`, `mistral/magistral-small-2506`, `mistral/mistral-large-2402`, `mistral/mistral-large-2407`, `mistral/mistral-large-2411`, `mistral/mistral-medium-2312`, `mistral/mistral-medium-2505`, `mistral/mistral-medium-2508`, `mistral/mistral-medium-3-1-2508`, `mistral/mistral-ocr-2505-completion`, `mistral/mistral-small-3-2-2506`, `mistral/open-codestral-mamba`, `mistral/open-mistral-7b`, `mistral/open-mistral-nemo-2407`, `mistral/open-mixtral-8x22b`, `mistral/open-mixtral-8x7b`, `mistral/pixtral-12b-2409`, `mistral/pixtral-large-2411`, `moonshot/kimi-k2-0711-preview`, `moonshot/kimi-k2-0905-preview`, `moonshot/kimi-k2-thinking`, `moonshot/kimi-k2-thinking-turbo`, `moonshot/kimi-k2-turbo-preview`, `moonshot/kimi-latest`, `moonshot/kimi-latest-128k`, `moonshot/kimi-latest-32k`, `moonshot/kimi-latest-8k`, `moonshot/kimi-thinking-preview`, `moonshot/moonshot-v1-128k-0430`, `moonshot/moonshot-v1-32k-0430`, `moonshot/moonshot-v1-8k-0430`, `o3-deep-research-2025-06-26`, `o4-mini-deep-research-2025-06-26`, `openrouter/anthropic/claude-opus-4`, `openrouter/deepseek/deepseek-v4-flash-0731:batch`, `openrouter/deepseek/deepseek-v4-flash-0731:free`, `openrouter/deepseek/deepseek-v4-flash-vision-exp:batch`, `openrouter/deepseek/deepseek-v4-pro-0813:batch`, `openrouter/google/gemini-2.0-flash-001`, `openrouter/kwaipilot/kat-coder-pro-v2`, `openrouter/meta/muse-glimmer-30b:batch`, `openrouter/minimax/minimax-m3:batch`, `openrouter/qwen/qwen3.5-9b:batch`, `openrouter/qwen/qwen3.8-2.4t-a95b:batch`, `openrouter/stealth/union-alpha`, `openrouter/thinkingmachines/inkling:batch`, `openrouter/z-ai/glm-5.2:batch`, `rerank-english-v2.0`, `rerank-multilingual-v2.0`, `sambanova/DeepSeek-R1-Distill-Llama-70B`, `sambanova/DeepSeek-V3-0324`, `sambanova/Llama-4-Scout-17B-16E-Instruct`, `sambanova/Meta-Llama-3.1-405B-Instruct`, `sambanova/Meta-Llama-3.1-8B-Instruct`, `sambanova/Meta-Llama-3.2-1B-Instruct`, `sambanova/Meta-Llama-3.2-3B-Instruct`, `sambanova/Meta-Llama-Guard-3-8B`, `sambanova/QwQ-32B`, `sambanova/Qwen2-Audio-7B-Instruct`, `sambanova/Qwen3-32B`, `scaleway/google/gemma-3-27b-it`, `scaleway/hcompany/holo2-30b-a3b`, `scaleway/mistralai/devstral-2-123b-instruct-2512`, `scaleway/mistralai/voxtral-small-24b-2507`, `standard/1024-x-1024/dall-e-3`, `standard/1024-x-1792/dall-e-3`, `standard/1792-x-1024/dall-e-3`, `text-moderation-007`, `text-moderation-latest`, `text-moderation-stable`, `together_ai/Qwen/Qwen3-235B-A22B-Instruct-2507-tput`, `together_ai/Qwen/Qwen3-235B-A22B-Thinking-2507`, `together_ai/Qwen/Qwen3-235B-A22B-fp8-tput`, `together_ai/deepseek-ai/DeepSeek-R1`, `together_ai/deepseek-ai/DeepSeek-R1-0528-tput`, `together_ai/deepseek-ai/DeepSeek-V4-Pro`, `together_ai/google/gemma-3n-E4B-it`, `together_ai/intfloat/multilingual-e5-large-instruct`, `together_ai/meta-llama/Llama-3.2-3B-Instruct-Turbo`, `together_ai/meta-llama/Llama-3.3-70B-Instruct-Turbo-Free`, `together_ai/meta-llama/Llama-4-Maverick-17B-128E-Instruct-FP8`, `together_ai/meta-llama/Llama-Guard-4-12B`, `together_ai/meta-llama/Meta-Llama-3.1-405B-Instruct-Turbo`, `together_ai/moonshotai/Kimi-K2-Instruct-0905`, `together_ai/moonshotai/Kimi-K2.5`, `together_ai/pearl-ai/gemma-4-31b-it`, `together_ai/thinkingmachines/Inkling-Small`, `us-gov.anthropic.claude-3-haiku-20240307-v1:0`, `us.amazon.nova-premier-v1:0`, `us.anthropic.claude-3-haiku-20240307-v1:0`, `us.anthropic.claude-3-sonnet-20240229-v1:0`, `vercel_ai_gateway/google/gemini-2.0-flash`, `vercel_ai_gateway/google/gemini-2.0-flash-lite`, `vertex_ai/claude-3-7-sonnet@20250219`, `vertex_ai/claude-opus-4`, `vertex_ai/claude-opus-4-1`, `vertex_ai/claude-opus-4-1@20250805`, `vertex_ai/claude-opus-4@20250514`, `vertex_ai/claude-sonnet-4`, `vertex_ai/claude-sonnet-4@20250514`, `vertex_ai/gemini-3.1-flash-live-preview`, `vertex_ai/gemini-robotics-er-2`, `vertex_ai/imagegeneration@006`, `vertex_ai/imagen-3.0-capability-001`, `vertex_ai/imagen-3.0-fast-generate-001`, `vertex_ai/imagen-3.0-generate-001`, `vertex_ai/imagen-3.0-generate-002`, `vertex_ai/imagen-4.0-fast-generate-001`, `vertex_ai/imagen-4.0-generate-001`, `vertex_ai/imagen-4.0-ultra-generate-001`, `wandb/MiniMaxAI/MiniMax-M2.5`, `wandb/Qwen/Qwen3-235B-A22B-Instruct-2507`, `wandb/Qwen/Qwen3-235B-A22B-Thinking-2507`, `wandb/Qwen/Qwen3-Coder-480B-A35B-Instruct`, `wandb/deepseek-ai/DeepSeek-R1-0528`, `wandb/deepseek-ai/DeepSeek-V3-0324`, `wandb/meta-llama/Llama-4-Scout-17B-16E-Instruct`, `wandb/microsoft/Phi-4-mini-instruct`, `wandb/moonshotai/Kimi-K2-Instruct`, `wandb/zai-org/GLM-4.5`, `xai/grok-3`, `xai/grok-3-beta`, `xai/grok-3-fast-beta`, `xai/grok-3-fast-latest`, `xai/grok-3-latest`, `xai/grok-3-mini`, `xai/grok-3-mini-beta`, `xai/grok-3-mini-fast`, `xai/grok-3-mini-fast-beta`, `xai/grok-3-mini-fast-latest`, `xai/grok-3-mini-latest`, `xai/grok-4`, `xai/grok-4-0709`, `xai/grok-4-1-fast`, `xai/grok-4-1-fast-non-reasoning`, `xai/grok-4-1-fast-non-reasoning-latest`, `xai/grok-4-1-fast-reasoning`, `xai/grok-4-1-fast-reasoning-latest`, `xai/grok-4-fast-non-reasoning`, `xai/grok-4-fast-reasoning`, `xai/grok-4-latest`
### New providers - Add Nadir intelligent-router provider (nadir/auto) - [PR #33227](https://github.com/BerriAI/litellm/pull/33227) - Add Eden AI provider across chat, Responses, Messages, embeddings, audio, images and video - [PR #41101](https://github.com/BerriAI/litellm/pull/41101) ### Amazon Bedrock - Strip body params the AWS endpoint rejects - [PR #31203](https://github.com/BerriAI/litellm/pull/31203) - Drop unsupported sampling params on converse reasoning models - [PR #39834](https://github.com/BerriAI/litellm/pull/39834) - Send s3BucketOwner on batch input and output data config - [PR #42262](https://github.com/BerriAI/litellm/pull/42262) - Add us.moonshotai.kimi-k3 pricing and fill the global Kimi K3 entry - [PR #42271](https://github.com/BerriAI/litellm/pull/42271) - Forward anthropic-beta headers verbatim on the Claude platform messages path - [PR #42275](https://github.com/BerriAI/litellm/pull/42275) - Keep batch S3 credentials out of chat requests and debug logs - [PR #42312](https://github.com/BerriAI/litellm/pull/42312) - Sign batch S3 requests with s3_access_key_id and s3_secret_access_key - [PR #42342](https://github.com/BerriAI/litellm/pull/42342) - Add bare moonshotai.kimi-k3 cost map entry - [PR #42363](https://github.com/BerriAI/litellm/pull/42363) - Send every Mantle beta in the anthropic-beta header on the bedrock/mantle route - [PR #42376](https://github.com/BerriAI/litellm/pull/42376) - Price bedrock/mantle/<model> deployments from the base model row - [PR #42402](https://github.com/BerriAI/litellm/pull/42402) - Treat blank AWS_S3_\* env vars as unset for batch jobs - [PR #42528](https://github.com/BerriAI/litellm/pull/42528) - Add Claude Opus 5.5 pricing and capabilities - [PR #42588](https://github.com/BerriAI/litellm/pull/42588) - Send json_schema as a forced tool on Claude Opus 4.7 and 4.8 Converse - [PR #42644](https://github.com/BerriAI/litellm/pull/42644) - Honour stream_chunk_size in Invoke streaming - [PR #42686](https://github.com/BerriAI/litellm/pull/42686) - Route unmapped openai family model ids to converse - [PR #42713](https://github.com/BerriAI/litellm/pull/42713) - Add gpt-6-sol and gpt-6-luna model pricing - [PR #42746](https://github.com/BerriAI/litellm/pull/42746) - Serve the OpenAI models on bedrock-runtime's native Responses API (internal copy of #38489) - [PR #42767](https://github.com/BerriAI/litellm/pull/42767) - Add bare openai.gpt-6-sol and openai.gpt-6-luna cost map rows - [PR #42798](https://github.com/BerriAI/litellm/pull/42798) - Add 17 aws-bedrock cost map rows from provider sync - [PR #42852](https://github.com/BerriAI/litellm/pull/42852) - Add gpt-5.4 and gpt-5.5 us and global inference profile pricing - [PR #42941](https://github.com/BerriAI/litellm/pull/42941) - Extrapolate global cris pricing for gpt-5.4 and gpt-5.5 - [PR #42971](https://github.com/BerriAI/litellm/pull/42971) - Map Anthropic batch row params the way real time does - [PR #43087](https://github.com/BerriAI/litellm/pull/43087) ### Anthropic - Forward safeguards and anthropic-beta unchanged on native /v1/messages - [PR #42152](https://github.com/BerriAI/litellm/pull/42152) - Forward Claude Code safeguards and dangerous-tool-use beta to Bedrock Invoke and Vertex on /v1/messages - [PR #42288](https://github.com/BerriAI/litellm/pull/42288) - Keep reasoning_effort a string for targets that stay on chat completions - [PR #42401](https://github.com/BerriAI/litellm/pull/42401) - Return 400 instead of 500 when a content list holds a bare string - [PR #42420](https://github.com/BerriAI/litellm/pull/42420) - Add Claude Opus 5.5 - [PR #42489](https://github.com/BerriAI/litellm/pull/42489) - Keep the replayed prefix byte-stable for preserved thinking on chat completions - [PR #42630](https://github.com/BerriAI/litellm/pull/42630) - Preserve MCP tool results in the non-Anthropic Messages bridge - [PR #42783](https://github.com/BerriAI/litellm/pull/42783) ### Azure - Bridge gpt-5.4+ function tools with reasoning to the Foundry Responses API - [PR #42041](https://github.com/BerriAI/litellm/pull/42041) - Propagate asyncio.CancelledError instead of raising a 500 - [PR #42295](https://github.com/BerriAI/litellm/pull/42295) - Add gpt-audio and gpt-realtime alias rows from the Azure model list - [PR #42981](https://github.com/BerriAI/litellm/pull/42981) - Update gpt-audio-mini and gpt-5-chat deprecation dates from the retirement schedule - [PR #43017](https://github.com/BerriAI/litellm/pull/43017) ### Bedrock Mantle - Serve /v1/messages for Claude models on Mantle's native Anthropic Messages API - [PR #42049](https://github.com/BerriAI/litellm/pull/42049) ### fal.ai - Add Seedance 2.5 / 2.0 video generation via fal queue API - [PR #41980](https://github.com/BerriAI/litellm/pull/41980) - Add gpt-image-2.5 flare/sunburst, flux/dev and image edits - [PR #42095](https://github.com/BerriAI/litellm/pull/42095) - Price images from the dimensions fal returns - [PR #42282](https://github.com/BerriAI/litellm/pull/42282) - Add MiniMax H3 text-to-video and reference-to-video - [PR #42286](https://github.com/BerriAI/litellm/pull/42286) - Surface fal errors in video status and content instead of completed and generic 500 - [PR #42306](https://github.com/BerriAI/litellm/pull/42306) - Add flux-lora-depth image edits and moondream3 chat completions - [PR #42334](https://github.com/BerriAI/litellm/pull/42334) - Price non-canonical image sizes from the nearest row and honour dump options - [PR #42336](https://github.com/BerriAI/litellm/pull/42336) - Add queue-only /fal_ai pass-through route with spend tracking - [PR #42360](https://github.com/BerriAI/litellm/pull/42360) - Handle seconds=auto and oversized sizes for minimax h3 videos - [PR #42504](https://github.com/BerriAI/litellm/pull/42504) - Align /fal_ai queue gate with the pricer and normalise resolution type - [PR #42505](https://github.com/BerriAI/litellm/pull/42505) - Reuse the status client and headers on the video result probe - [PR #42511](https://github.com/BerriAI/litellm/pull/42511) - Honour global api_base for image generation and reject non-string reasoning_effort with 400 - [PR #42512](https://github.com/BerriAI/litellm/pull/42512) - Price nano-banana-2 and nano-banana-pro image generations by resolution - [PR #43101](https://github.com/BerriAI/litellm/pull/43101) ### Fireworks AI - Route firerouter short names and bill pass-through legs at the routed model's rates - [PR #42814](https://github.com/BerriAI/litellm/pull/42814) ### Gemini and Vertex AI - Forward response schema and tool parameters through the generateContent adapter - [PR #42067](https://github.com/BerriAI/litellm/pull/42067) - Simplify model version check - [PR #42465](https://github.com/BerriAI/litellm/pull/42465) - Add gemini-3.8-flash-tts and gemini-3.8-flash-lite-tts prices - [PR #42752](https://github.com/BerriAI/litellm/pull/42752) - Native batch JSONL passthrough with cost tracking - [PR #42810](https://github.com/BerriAI/litellm/pull/42810) - Add deprecation dates for retired claude 3 and jamba 1.5 partner models - [PR #42867](https://github.com/BerriAI/litellm/pull/42867) - Add gemini-3.8-flash-cyber pricing - [PR #42879](https://github.com/BerriAI/litellm/pull/42879) - Keep batch output_file_id null until Vertex reports outputInfo - [PR #43030](https://github.com/BerriAI/litellm/pull/43030) - Translate /v1/responses batch rows through the Responses-to-Chat bridge - [PR #43042](https://github.com/BerriAI/litellm/pull/43042) - Surface the Gemma container's own error inside a 200 :predict response - [PR #43075](https://github.com/BerriAI/litellm/pull/43075) - Stop advertising OpenAI platform-only params on Gemma and Llama routes - [PR #43079](https://github.com/BerriAI/litellm/pull/43079) - Return chunk content, extractive text, and structData from search_api vector store hits - [PR #43100](https://github.com/BerriAI/litellm/pull/43100) ### GitHub Copilot and ChatGPT - Answer get_api_base for github_copilot and chatgpt without running the login flow - [PR #42602](https://github.com/BerriAI/litellm/pull/42602) ### Ollama - Send PNG and JPEG images without requiring Pillow - [PR #41979](https://github.com/BerriAI/litellm/pull/41979) - Read the JSON thinking field on non-streaming completions - [PR #42838](https://github.com/BerriAI/litellm/pull/42838) ### OpenAI - Forward non-enum reasoning_effort through the Responses bridge instead of dropping it - [PR #42452](https://github.com/BerriAI/litellm/pull/42452) - Add GPT-6 Sol and GPT-6 Luna - [PR #42515](https://github.com/BerriAI/litellm/pull/42515) ### OpenRouter - Price typesafe/jev-1.13 and add an openrouter decisions pass-through - [PR #42301](https://github.com/BerriAI/litellm/pull/42301) - Remove the retired stealth/union-alpha model from the cost map - [PR #42386](https://github.com/BerriAI/litellm/pull/42386) ### Qianwen AI Platform - Rename the mainland China brand to Qianwen AI Platform - [PR #42284](https://github.com/BerriAI/litellm/pull/42284) ### Together AI - Backfill deprecation_date from Together deprecation history - [PR #43135](https://github.com/BerriAI/litellm/pull/43135) ### xAI - Add grok-4.7 to the cost map - [PR #42264](https://github.com/BerriAI/litellm/pull/42264) - Accept max_completion_tokens as a supported param - [PR #42353](https://github.com/BerriAI/litellm/pull/42353) ### Xiaomi MiMo - Add mimo-v2.6-pro and mimo-v2.6-flash cost map rows with live e2e coverage - [PR #42362](https://github.com/BerriAI/litellm/pull/42362) ### Model catalog and pricing - **New and updated catalog rows** - Add MAI-Image-2.5-Pro pricing, fix Fireworks/Together entries, absorb verified open registry PRs, add Groq deprecation and Bedrock regional Qwen3 Next pricing - [PR #34941](https://github.com/BerriAI/litellm/pull/34941) - Add xai grok-4.20 aliases and image token prices from /v1/language-models - [PR #42384](https://github.com/BerriAI/litellm/pull/42384) - Add Claude Opus 5.5 for Vertex AI and Azure AI - [PR #42599](https://github.com/BerriAI/litellm/pull/42599) - Add openrouter/aion-labs/aion-3.5-mini pricing - [PR #42743](https://github.com/BerriAI/litellm/pull/42743) - Add Azure Foundry pricing for gpt-6-sol and gpt-6-luna - [PR #42747](https://github.com/BerriAI/litellm/pull/42747) - Add openrouter/stealth/space-bunny-alpha - [PR #42759](https://github.com/BerriAI/litellm/pull/42759) - Add baseten/zai-org/GLM-5.3-Fast pricing - [PR #42764](https://github.com/BerriAI/litellm/pull/42764) - Add together_ai/together/Tev1-4B-experimental - [PR #42807](https://github.com/BerriAI/litellm/pull/42807) - Add gemini preview aliases and deep research 04-2026 rows - [PR #42833](https://github.com/BerriAI/litellm/pull/42833) - Add openai chat-latest, codex and deep-research rows from the model docs - [PR #42834](https://github.com/BerriAI/litellm/pull/42834) - Add vertex ai llama 3.3 70b, veo 2/3, virtual try-on and 2.5 tts rows - [PR #42837](https://github.com/BerriAI/litellm/pull/42837) - Add openrouter/openai/gpt-oss-120b:batch from the OpenRouter models API - [PR #42847](https://github.com/BerriAI/litellm/pull/42847) - Add gemini lyria-realtime-exp row inherited from lyria-3.5 - [PR #42848](https://github.com/BerriAI/litellm/pull/42848) - Add 39 together_ai chat rows priced by the Together models API - [PR #42851](https://github.com/BerriAI/litellm/pull/42851) - Add retired azure gpt-5 chat and o1-preview data zone rows - [PR #42878](https://github.com/BerriAI/litellm/pull/42878) - Add vertex_ai/gemini-3.8-live - [PR #42891](https://github.com/BerriAI/litellm/pull/42891) - Add wandb DeepSeek-V4.1-Flash and gemma-4-26B-A4B-it - [PR #42924](https://github.com/BerriAI/litellm/pull/42924) - Add azure realtime, audio and partner model rows - [PR #42954](https://github.com/BerriAI/litellm/pull/42954) - Sync azure models, add MAI-Image-2.6, deepseek-v4.1-flash, muse-spark-1.3 - [PR #42970](https://github.com/BerriAI/litellm/pull/42970) - Add supports_reasoning to azure/eu/gpt-6-astra - [PR #42989](https://github.com/BerriAI/litellm/pull/42989) - Add fireworks glm-5p3 US-only rows and kimi-k3-us priority prices - [PR #43092](https://github.com/BerriAI/litellm/pull/43092) - Add fireworks deepseek-v4p1-flash US-only rows - [PR #43097](https://github.com/BerriAI/litellm/pull/43097) - **Price corrections** - Registry audit 2026-09-22, absorb open pricing PRs - [PR #42543](https://github.com/BerriAI/litellm/pull/42543) - Drop the unpublished cached rate from the Gemini Live preview entries - [PR #42651](https://github.com/BerriAI/litellm/pull/42651) - Align bedrock_mantle/openai.gpt-daybreak-blue-5.6-sol with its Bedrock model card - [PR #42672](https://github.com/BerriAI/litellm/pull/42672) - Align regional Bedrock Mistral Large 24.02 keys with the AWS pricing page - [PR #42684](https://github.com/BerriAI/litellm/pull/42684) - Registry audit 2026-09-23, in-region Bedrock Claude prices - [PR #42779](https://github.com/BerriAI/litellm/pull/42779) - Sync openrouter prices from the models API - [PR #42832](https://github.com/BerriAI/litellm/pull/42832) - Bedrock bare Claude ids priced at the Global SKU (aws-bedrock sync) - [PR #42875](https://github.com/BerriAI/litellm/pull/42875) - Correct gemini robotics er 2 preview audio input price - [PR #42877](https://github.com/BerriAI/litellm/pull/42877) - Halve openrouter deepseek-v4-flash-0731 output price - [PR #42881](https://github.com/BerriAI/litellm/pull/42881) - Correct fireworks kimi k3 us pricing to the published rate - [PR #42884](https://github.com/BerriAI/litellm/pull/42884) - Sync openrouter prices and add fireworks ember-1 - [PR #42889](https://github.com/BerriAI/litellm/pull/42889) - Source and chat completions endpoint for bedrock mantle gpt-5.4 and gpt-5.5 - [PR #42890](https://github.com/BerriAI/litellm/pull/42890) - Source for bedrock mantle gpt-5.6 luna, sol, terra and grok-4.6 - [PR #42898](https://github.com/BerriAI/litellm/pull/42898) - Update openrouter kimi-k2.7-code input price - [PR #42932](https://github.com/BerriAI/litellm/pull/42932) - Sync vertex-ai rows (gemma 4 maas cache price, chirp_2) - [PR #42942](https://github.com/BerriAI/litellm/pull/42942) - Retirement dates, chatgpt reasoning flags, bing pricing, bedrock mantle and mythos, azure gpt-5.6 alias, anthropic batch rates, new nebius, openrouter and xai rows - [PR #42951](https://github.com/BerriAI/litellm/pull/42951) - Sync openrouter prices for deepseek v4 and glm-5.3 - [PR #42952](https://github.com/BerriAI/litellm/pull/42952) - Add batch prices for vertex gemini-3.8-flash-cyber - [PR #42953](https://github.com/BerriAI/litellm/pull/42953) - Sync openrouter deepseek-v4-pro prices - [PR #42956](https://github.com/BerriAI/litellm/pull/42956) - Sync openrouter deepseek-v4-pro prices - [PR #42964](https://github.com/BerriAI/litellm/pull/42964) - Sync openrouter deepseek-v4-pro prices - [PR #42969](https://github.com/BerriAI/litellm/pull/42969) - Sync openrouter deepseek v4 flash, v4 pro and v4.1 flash prices - [PR #42974](https://github.com/BerriAI/litellm/pull/42974) - Sync openrouter deepseek-v4-pro prices - [PR #42978](https://github.com/BerriAI/litellm/pull/42978) - Add vertex ai priority audio input prices for gemini flash rows - [PR #42980](https://github.com/BerriAI/litellm/pull/42980) - Align openrouter deepseek-v4-pro cache hit price with cache read - [PR #42985](https://github.com/BerriAI/litellm/pull/42985) - Drop stale cache_hit field from openrouter deepseek-v4-pro - [PR #42994](https://github.com/BerriAI/litellm/pull/42994) - Add vertex cache read, batch and above 200k prices for gemini image preview rows - [PR #42995](https://github.com/BerriAI/litellm/pull/42995) - Add video and reasoning output prices to vertex gemini-omni-1.1-flash - [PR #43036](https://github.com/BerriAI/litellm/pull/43036) - Drop the priority input price from vertex gemini-2.5-flash-image - [PR #43050](https://github.com/BerriAI/litellm/pull/43050) - Add vertex priority prices for gemini-3-pro-image-preview and batch price for gemini-embedding-001 - [PR #43069](https://github.com/BerriAI/litellm/pull/43069) - Add video input price to gemini 3.8 live rows - [PR #43073](https://github.com/BerriAI/litellm/pull/43073) - Drop cache read price from vertex gemini-2.5-flash-image - [PR #43078](https://github.com/BerriAI/litellm/pull/43078) - Sync openrouter deepseek v4 and glm-5.3-flash prices, add mistral-large-2512 - [PR #43090](https://github.com/BerriAI/litellm/pull/43090) - Sync gemini priority, flex and video token prices from the Gemini API pricing page - [PR #43091](https://github.com/BerriAI/litellm/pull/43091) - Add gemini tts batch output prices from the Gemini API pricing page - [PR #43103](https://github.com/BerriAI/litellm/pull/43103) - Add openai cached image input prices from the pricing page - [PR #43143](https://github.com/BerriAI/litellm/pull/43143) - **Deprecation and retirement dates** - Add azure_ai gpt-image-2 and groq llama-guard-3-8b deprecation dates - [PR #42738](https://github.com/BerriAI/litellm/pull/42738) - Add deprecation_date to claude-mythos-preview from the Anthropic deprecations page - [PR #42845](https://github.com/BerriAI/litellm/pull/42845) - Add the sora-2-pro shutdown date to the sora-2-pro-high-res rows - [PR #42846](https://github.com/BerriAI/litellm/pull/42846) - Add fireworks deprecation dates for kimi k2.6 fast, kimi k2.7 code fast and glm 5.2 fast us - [PR #42849](https://github.com/BerriAI/litellm/pull/42849) - Add the June 1, 2026 retirement date to the vertex_ai gemini-2.0-flash rows - [PR #42850](https://github.com/BerriAI/litellm/pull/42850) - Add fireworks deprecation date for glm 5.2 fast serverless rows - [PR #42866](https://github.com/BerriAI/litellm/pull/42866) - Add openai deprecation date for gpt-5-chat-latest and gpt-5-chat - [PR #42872](https://github.com/BerriAI/litellm/pull/42872) - Add azure gpt-4o-mini-transcribe and gpt-4o-mini-tts deprecation dates - [PR #42873](https://github.com/BerriAI/litellm/pull/42873) - Add fireworks 2026-09-25 deprecation dates for glm 5.2, kimi k2.6, kimi k2.7 code, deepseek v4 and muse glimmer rows - [PR #42874](https://github.com/BerriAI/litellm/pull/42874) - Sync vertex-ai deprecation dates from Vertex model lifecycle pages - [PR #42882](https://github.com/BerriAI/litellm/pull/42882) - Add azure gpt-realtime-mini deprecation date - [PR #42883](https://github.com/BerriAI/litellm/pull/42883) - Add azure gpt-realtime-mini-2025-10-06 deprecation date - [PR #42885](https://github.com/BerriAI/litellm/pull/42885) - Add azure deprecation dates for gpt-6 and gpt-realtime-whisper - [PR #42897](https://github.com/BerriAI/litellm/pull/42897) - Add azure deprecation dates for regional gpt-6 rows - [PR #42933](https://github.com/BerriAI/litellm/pull/42933) - Update azure gpt-4.1-nano and gpt-4o-2024-05-13 retirement dates - [PR #42947](https://github.com/BerriAI/litellm/pull/42947) - Take the later azure deprecation date for gpt-4.1-nano, gpt-4o-transcribe and gpt-realtime-2.1 - [PR #42986](https://github.com/BerriAI/litellm/pull/42986) - Add the Vertex shutdown date to gemini-2.5-flash-native-audio - [PR #43024](https://github.com/BerriAI/litellm/pull/43024) - Add azure deprecation dates from the Models API for five realtime and transcribe rows - [PR #43037](https://github.com/BerriAI/litellm/pull/43037) - Move azure gpt-realtime-2.1-mini deprecation date to the later Models API date - [PR #43058](https://github.com/BerriAI/litellm/pull/43058) - Add openai deprecation dates from the deprecations page - [PR #43102](https://github.com/BerriAI/litellm/pull/43102) - Add azure retirement dates from the retired Foundry models page - [PR #43104](https://github.com/BerriAI/litellm/pull/43104) - Add computer-use-preview deprecation date from the openai deprecations page - [PR #43116](https://github.com/BerriAI/litellm/pull/43116) - Add azure retirement dates for command-r-plus and gpt-4 - [PR #43117](https://github.com/BerriAI/litellm/pull/43117) - Add together-ai deprecation dates for gpt-oss-20b and gemma-4-31B-it - [PR #43127](https://github.com/BerriAI/litellm/pull/43127) - **Removals and cleanup** - Remove models past their deprecation date - [PR #42435](https://github.com/BerriAI/litellm/pull/42435) - Remove retired models flagged by the provider sync - [PR #42521](https://github.com/BerriAI/litellm/pull/42521) - Drop duplicate cache_read_input_token_cost_batches keys from the price maps - [PR #42623](https://github.com/BerriAI/litellm/pull/42623) - **Automated provider price syncs** (78 PRs: OpenRouter 51, AWS Bedrock 11, OpenAI 3, Baseten 3, Vertex AI 3, Azure 2, Google Gemini 2, Together AI 1, Fireworks AI 1, xAI 1) - [PR #42077](https://github.com/BerriAI/litellm/pull/42077), [PR #42082](https://github.com/BerriAI/litellm/pull/42082), [PR #42089](https://github.com/BerriAI/litellm/pull/42089), [PR #42092](https://github.com/BerriAI/litellm/pull/42092), [PR #42096](https://github.com/BerriAI/litellm/pull/42096), [PR #42154](https://github.com/BerriAI/litellm/pull/42154), [PR #42155](https://github.com/BerriAI/litellm/pull/42155), [PR #42157](https://github.com/BerriAI/litellm/pull/42157), [PR #42162](https://github.com/BerriAI/litellm/pull/42162), [PR #42163](https://github.com/BerriAI/litellm/pull/42163), [PR #42164](https://github.com/BerriAI/litellm/pull/42164), [PR #42166](https://github.com/BerriAI/litellm/pull/42166), [PR #42168](https://github.com/BerriAI/litellm/pull/42168), [PR #42169](https://github.com/BerriAI/litellm/pull/42169), [PR #42175](https://github.com/BerriAI/litellm/pull/42175), [PR #42178](https://github.com/BerriAI/litellm/pull/42178), [PR #42179](https://github.com/BerriAI/litellm/pull/42179), [PR #42182](https://github.com/BerriAI/litellm/pull/42182), [PR #42184](https://github.com/BerriAI/litellm/pull/42184), [PR #42187](https://github.com/BerriAI/litellm/pull/42187), [PR #42192](https://github.com/BerriAI/litellm/pull/42192), [PR #42227](https://github.com/BerriAI/litellm/pull/42227), [PR #42234](https://github.com/BerriAI/litellm/pull/42234), [PR #42240](https://github.com/BerriAI/litellm/pull/42240), [PR #42243](https://github.com/BerriAI/litellm/pull/42243), [PR #42246](https://github.com/BerriAI/litellm/pull/42246), [PR #42249](https://github.com/BerriAI/litellm/pull/42249), [PR #42250](https://github.com/BerriAI/litellm/pull/42250), [PR #42251](https://github.com/BerriAI/litellm/pull/42251), [PR #42253](https://github.com/BerriAI/litellm/pull/42253), [PR #42254](https://github.com/BerriAI/litellm/pull/42254), [PR #42258](https://github.com/BerriAI/litellm/pull/42258), [PR #42261](https://github.com/BerriAI/litellm/pull/42261), [PR #42265](https://github.com/BerriAI/litellm/pull/42265), [PR #42269](https://github.com/BerriAI/litellm/pull/42269), [PR #42270](https://github.com/BerriAI/litellm/pull/42270), [PR #42279](https://github.com/BerriAI/litellm/pull/42279), [PR #42280](https://github.com/BerriAI/litellm/pull/42280), [PR #42289](https://github.com/BerriAI/litellm/pull/42289), [PR #42290](https://github.com/BerriAI/litellm/pull/42290), [PR #42297](https://github.com/BerriAI/litellm/pull/42297), [PR #42298](https://github.com/BerriAI/litellm/pull/42298), [PR #42305](https://github.com/BerriAI/litellm/pull/42305), [PR #42320](https://github.com/BerriAI/litellm/pull/42320), [PR #42333](https://github.com/BerriAI/litellm/pull/42333), [PR #42337](https://github.com/BerriAI/litellm/pull/42337), [PR #42338](https://github.com/BerriAI/litellm/pull/42338), [PR #42349](https://github.com/BerriAI/litellm/pull/42349), [PR #42357](https://github.com/BerriAI/litellm/pull/42357), [PR #42365](https://github.com/BerriAI/litellm/pull/42365), [PR #42371](https://github.com/BerriAI/litellm/pull/42371), [PR #42377](https://github.com/BerriAI/litellm/pull/42377), [PR #42381](https://github.com/BerriAI/litellm/pull/42381), [PR #42407](https://github.com/BerriAI/litellm/pull/42407), [PR #42418](https://github.com/BerriAI/litellm/pull/42418), [PR #42438](https://github.com/BerriAI/litellm/pull/42438), [PR #42485](https://github.com/BerriAI/litellm/pull/42485), [PR #42500](https://github.com/BerriAI/litellm/pull/42500), [PR #42501](https://github.com/BerriAI/litellm/pull/42501), [PR #42502](https://github.com/BerriAI/litellm/pull/42502), [PR #42557](https://github.com/BerriAI/litellm/pull/42557), [PR #42558](https://github.com/BerriAI/litellm/pull/42558), [PR #42577](https://github.com/BerriAI/litellm/pull/42577), [PR #42578](https://github.com/BerriAI/litellm/pull/42578), [PR #42589](https://github.com/BerriAI/litellm/pull/42589), [PR #42590](https://github.com/BerriAI/litellm/pull/42590), [PR #42591](https://github.com/BerriAI/litellm/pull/42591), [PR #42592](https://github.com/BerriAI/litellm/pull/42592), [PR #42632](https://github.com/BerriAI/litellm/pull/42632), [PR #42642](https://github.com/BerriAI/litellm/pull/42642), [PR #42648](https://github.com/BerriAI/litellm/pull/42648), [PR #42673](https://github.com/BerriAI/litellm/pull/42673), [PR #42677](https://github.com/BerriAI/litellm/pull/42677), [PR #42680](https://github.com/BerriAI/litellm/pull/42680), [PR #42685](https://github.com/BerriAI/litellm/pull/42685), [PR #42756](https://github.com/BerriAI/litellm/pull/42756), [PR #42771](https://github.com/BerriAI/litellm/pull/42771), [PR #42806](https://github.com/BerriAI/litellm/pull/42806) ## LLM API Endpoints ### Responses API - Stream one lifecycle across MCP auto-execute rounds - [PR #40121](https://github.com/BerriAI/litellm/pull/40121) - Stop agentic follow-up from passing request params twice - [PR #41560](https://github.com/BerriAI/litellm/pull/41560) - Patch custom_tool_call_output in place on guardrail write-back - [PR #41561](https://github.com/BerriAI/litellm/pull/41561) - Forward safety_identifier through the chat completion bridge - [PR #42193](https://github.com/BerriAI/litellm/pull/42193) - Build follow-up kwargs in one place so no executor can repeat a request param - [PR #42307](https://github.com/BerriAI/litellm/pull/42307) - Drop client_metadata and merge system messages for Databricks chat-only models - [PR #42390](https://github.com/BerriAI/litellm/pull/42390) ### Agents and Agent-to-Agent - Attach access groups to agents and enforce them for models, MCP servers and agent calls - [PR #41634](https://github.com/BerriAI/litellm/pull/41634) - Send message/stream for Bedrock AgentCore streaming requests - [PR #42239](https://github.com/BerriAI/litellm/pull/42239) - Add optional per-agent kill switch webhook - [PR #42841](https://github.com/BerriAI/litellm/pull/42841) ### OCR - Remove the Python OCR execution path and require the Rust route - [PR #43081](https://github.com/BerriAI/litellm/pull/43081) ### Realtime and Audio - Surface an upstream handshake refusal as an error event and policy close - [PR #42388](https://github.com/BerriAI/litellm/pull/42388) ### Vector Stores and Search - Keep config-defined vector stores listed and read-only - [PR #42574](https://github.com/BerriAI/litellm/pull/42574) ### Pass-through endpoints - Add TinyFish Agent API passthrough with per-step billing - [PR #41099](https://github.com/BerriAI/litellm/pull/41099) - Log upstream 4xx/5xx error bodies and carry them into the failure hook - [PR #42695](https://github.com/BerriAI/litellm/pull/42695) ### General - Stop /utils/transform_request from calling the provider and blocking the event loop - [PR #33954](https://github.com/BerriAI/litellm/pull/33954) - Prefilled GitHub issue link on unmapped internal errors - [PR #42065](https://github.com/BerriAI/litellm/pull/42065) - Keep an explicit provider prompt_tokens=0 or completion_tokens=0 in streamed usage - [PR #42323](https://github.com/BerriAI/litellm/pull/42323) - Let a later usage event zero out stale cache counts (#40736) - [PR #42330](https://github.com/BerriAI/litellm/pull/42330) - Opt-in litellm_call_id in JSON error bodies - [PR #42391](https://github.com/BerriAI/litellm/pull/42391) - Add stream and safe config flags to the bug report link - [PR #42428](https://github.com/BerriAI/litellm/pull/42428) - Stop a nested additional_drop_params entry from crashing openai-compatible calls - [PR #42492](https://github.com/BerriAI/litellm/pull/42492) - Point the blocked-address remediation at litellm_settings - [PR #42508](https://github.com/BerriAI/litellm/pull/42508) - Isolate callback errors in async_post_call_success_deployment_hook - [PR #42535](https://github.com/BerriAI/litellm/pull/42535) - Repair seven regressions caught by CircleCI on main - [PR #42640](https://github.com/BerriAI/litellm/pull/42640) - Stop stream_chunk_size reaching provider request bodies - [PR #42664](https://github.com/BerriAI/litellm/pull/42664) ## Management Endpoints / UI ### Admin UI - Surface the owner's user budget on keys without their own budget - [PR #38220](https://github.com/BerriAI/litellm/pull/38220) - Add upgrade banner with latest release changelog stats - [PR #40429](https://github.com/BerriAI/litellm/pull/40429) - Let the Create Key user picker find users by user_id, not just email - [PR #41687](https://github.com/BerriAI/litellm/pull/41687) - Add internal-user savings and auto-router usage - [PR #42026](https://github.com/BerriAI/litellm/pull/42026) - Show prompt caching requests and net savings - [PR #42055](https://github.com/BerriAI/litellm/pull/42055) - Show Capability and FUSE v2 routing forecasts - [PR #42057](https://github.com/BerriAI/litellm/pull/42057) - Expose remaining complexity router advanced settings - [PR #42293](https://github.com/BerriAI/litellm/pull/42293) - Add per-user breakdown to team usage export - [PR #42367](https://github.com/BerriAI/litellm/pull/42367) - Rename reminder markers to Ignore Custom Tags - [PR #42370](https://github.com/BerriAI/litellm/pull/42370) - Add native LiteAdmin assistant - [PR #42443](https://github.com/BerriAI/litellm/pull/42443) - Add span type filter to request logs - [PR #42491](https://github.com/BerriAI/litellm/pull/42491) - Simplify auto-router setup and clarify feature limits - [PR #42625](https://github.com/BerriAI/litellm/pull/42625) - Show ten prompt caching requests per page - [PR #42638](https://github.com/BerriAI/litellm/pull/42638) - Prefer native providers in auto-router presets - [PR #42639](https://github.com/BerriAI/litellm/pull/42639) - Keep per-user MCP credentials updatable and clearable after setup - [PR #42652](https://github.com/BerriAI/litellm/pull/42652) - Keep untouched stored auto-router booleans and reminder marker casing on save - [PR #42703](https://github.com/BerriAI/litellm/pull/42703) - Hide LiteAdmin in Playground and add admin preference - [PR #42755](https://github.com/BerriAI/litellm/pull/42755) - Make the audit log detail drawer wider and resizable - [PR #42808](https://github.com/BerriAI/litellm/pull/42808) - Offer reset of custom member budgets when team default changes - [PR #42835](https://github.com/BerriAI/litellm/pull/42835) - Configure prompt caching request rows per page - [PR #42842](https://github.com/BerriAI/litellm/pull/42842) - Explain unbackfilled key lifetime spend and ship a backfill script - [PR #42967](https://github.com/BerriAI/litellm/pull/42967) - Stop following streamed tokens, add jump to bottom button - [PR #42968](https://github.com/BerriAI/litellm/pull/42968) - Pass is_proxy_admin for proxy admins on the models page team drill-in - [PR #43003](https://github.com/BerriAI/litellm/pull/43003) - Group cost optimization cache leakage by model group - [PR #43008](https://github.com/BerriAI/litellm/pull/43008) ### Keys, teams and authentication - Invalidate cached object permissions on key update - [PR #36719](https://github.com/BerriAI/litellm/pull/36719) - Accept token_id as an alternative to the plaintext key - [PR #39578](https://github.com/BerriAI/litellm/pull/39578) - Count team unified access group MCP servers when validating key MCP grants - [PR #41231](https://github.com/BerriAI/litellm/pull/41231) - Use the user's own budget as the ceiling for UI session personal keys - [PR #41588](https://github.com/BerriAI/litellm/pull/41588) - Show whether a member follows the team default budget and allow resetting to it - [PR #41906](https://github.com/BerriAI/litellm/pull/41906) - Refuse to start with an unset, empty, or publicly known master key - [PR #42019](https://github.com/BerriAI/litellm/pull/42019) - Fail closed when the team membership lookup hits a db outage - [PR #42036](https://github.com/BerriAI/litellm/pull/42036) - Reject deactivated JWT users and refresh cached status - [PR #42064](https://github.com/BerriAI/litellm/pull/42064) - Breached password detection, self-service change-password and forced password reset - [PR #42278](https://github.com/BerriAI/litellm/pull/42278) - Evict the cached user row when SCIM or /user/delete removes a user - [PR #42315](https://github.com/BerriAI/litellm/pull/42315) - Fail closed when the JWT single-team fallback or compact editor membership read hits a DB outage - [PR #42344](https://github.com/BerriAI/litellm/pull/42344) - Let jwt team_allowed_routes paths grant auth=true passthrough - [PR #42346](https://github.com/BerriAI/litellm/pull/42346) - Surface a database outage from the user read as 503 no_db_connection - [PR #42399](https://github.com/BerriAI/litellm/pull/42399) - Answer 503 no_db_connection on management routes when the caller's user read hits a database outage - [PR #42410](https://github.com/BerriAI/litellm/pull/42410) - Enforce disable_custom_api_keys from general_settings - [PR #42437](https://github.com/BerriAI/litellm/pull/42437) - Write key deleted audit logs for cascade and alias key deletions - [PR #42446](https://github.com/BerriAI/litellm/pull/42446) - Revoke UI session tokens on logout and password change - [PR #42463](https://github.com/BerriAI/litellm/pull/42463) - Configurable key_alias_pattern for key generate, update, and regenerate - [PR #42553](https://github.com/BerriAI/litellm/pull/42553) - Let team admins update member key budgets when enabled - [PR #42555](https://github.com/BerriAI/litellm/pull/42555) - Authorize key model aliases the same way as team aliases - [PR #43049](https://github.com/BerriAI/litellm/pull/43049) ### Proxy configuration - Keep config-defined deployments when a config read returns no model_list - [PR #41505](https://github.com/BerriAI/litellm/pull/41505) - Report the source of alerting, UI and router settings on read - [PR #41788](https://github.com/BerriAI/litellm/pull/41788) - Drop cost-map metadata echoed back on model save - [PR #41944](https://github.com/BerriAI/litellm/pull/41944) - Remove the dead telemetry flag from the SDK, proxy CLI and configs - [PR #42071](https://github.com/BerriAI/litellm/pull/42071) - Add GET /utils/model_info to look up cost map info for unregistered models - [PR #42121](https://github.com/BerriAI/litellm/pull/42121) - Detach stored credential when model editor selects None - [PR #42291](https://github.com/BerriAI/litellm/pull/42291) - Admin-only /debug/report sharing the bug report environment - [PR #42440](https://github.com/BerriAI/litellm/pull/42440) - Never render credential-bearing config keys in the bug report - [PR #42493](https://github.com/BerriAI/litellm/pull/42493) - Make the lazy OpenAPI snapshot byte-identical on every Python version - [PR #42519](https://github.com/BerriAI/litellm/pull/42519) - Validate model credential name only when it changes - [PR #42701](https://github.com/BerriAI/litellm/pull/42701) - Document request body and response schemas for the Responses API in OpenAPI - [PR #42802](https://github.com/BerriAI/litellm/pull/42802) - Honor model_info.discoverable on the model listing endpoints - [PR #42825](https://github.com/BerriAI/litellm/pull/42825) - List key and team model aliases in GET /v1/models - [PR #42908](https://github.com/BerriAI/litellm/pull/42908) - Let callbacks filter the model listing routes per caller - [PR #43027](https://github.com/BerriAI/litellm/pull/43027) ### CLI and coding agents - Add --validate_config dry-run flag - [PR #41705](https://github.com/BerriAI/litellm/pull/41705) - Preserve newer installed status lines during setup - [PR #42356](https://github.com/BerriAI/litellm/pull/42356) - Import proxy_server once on script-style boot - [PR #42584](https://github.com/BerriAI/litellm/pull/42584) ### Deployment - Render a fixed replicaCount on componentized deployments when HPA is disabled - [PR #42207](https://github.com/BerriAI/litellm/pull/42207) - Add a quickstart compose file served from the product repo - [PR #42326](https://github.com/BerriAI/litellm/pull/42326) - Expose /api/event_logging/batch on the gateway allowlist - [PR #42572](https://github.com/BerriAI/litellm/pull/42572) - Bump wolfi-base digest to pick up glibc 2.44-r6 - [PR #42643](https://github.com/BerriAI/litellm/pull/42643) ### Terraform - Expose server_metadata on litellm_key so undeclared metadata is visible - [PR #42453](https://github.com/BerriAI/litellm/pull/42453) - Add display_name to litellm_model resource and model data sources - [PR #42987](https://github.com/BerriAI/litellm/pull/42987) ## AI Integrations ### Guardrails - Scan each choice's tool-call arguments apart on n>1 streams and log why a rewrite was discarded - [PR #40986](https://github.com/BerriAI/litellm/pull/40986) - Straiker guardrail speaks the v3 platform API (/api/v3/detect) - [PR #41880](https://github.com/BerriAI/litellm/pull/41880) - Add default fallback policy attachments - [PR #42119](https://github.com/BerriAI/litellm/pull/42119) - Opt-in include_guardrail_response returns guardrail_information in the response - [PR #42327](https://github.com/BerriAI/litellm/pull/42327) - Mask PII in streaming /v1/messages output - [PR #42351](https://github.com/BerriAI/litellm/pull/42351) - Run key-attached guardrails on /v1/videos - [PR #42354](https://github.com/BerriAI/litellm/pull/42354) - Store the masked output in spend logs when Presidio masks the response - [PR #42441](https://github.com/BerriAI/litellm/pull/42441) - Keep inherited parent guardrails when a child policy condition misses - [PR #42548](https://github.com/BerriAI/litellm/pull/42548) - Gate disable_global_guardrails on keys and teams to proxy admins - [PR #42699](https://github.com/BerriAI/litellm/pull/42699) - Stream non-Anthropic raw SSE through the post_call hook unbuffered - [PR #42777](https://github.com/BerriAI/litellm/pull/42777) - Keep deployment labels on cache-hit post_call guardrail rejections - [PR #42780](https://github.com/BerriAI/litellm/pull/42780) - Mask streamed /v1/messages output when the first upstream read is a keepalive, a data-less ping, or a split utf8 character - [PR #43023](https://github.com/BerriAI/litellm/pull/43023) ### Logging and observability - Migrate the sdk callback to langfuse v4. The Docker image now ships `langfuse>=4.7,<5`. If you install the SDK yourself, upgrade from `langfuse` 2.x or 3.x, which the logger now rejects at startup. Traces go through Langfuse's OpenTelemetry (OTLP) ingestion, so a self-hosted Langfuse server must accept OTLP, and trace ids and span nesting differ from the v2 integration - [PR #36741](https://github.com/BerriAI/litellm/pull/36741) - Replace colons in generated log filenames - [PR #40452](https://github.com/BerriAI/litellm/pull/40452) - Attribute provider and model_info on pre_call_hook rejections - [PR #41077](https://github.com/BerriAI/litellm/pull/41077) - Bound concurrent S3 uploads per flush and add opt-in JSONL batch files - [PR #41258](https://github.com/BerriAI/litellm/pull/41258) - Keep intercepted searches under the parent request's session and trace - [PR #41711](https://github.com/BerriAI/litellm/pull/41711) - Add normalized_error cluster key to error_information - [PR #41715](https://github.com/BerriAI/litellm/pull/41715) - Honor SSL_CERT_FILE and ssl_verify in OTLP HTTP exporters - [PR #42106](https://github.com/BerriAI/litellm/pull/42106) - Apply DB-stored callback redaction settings before logger init - [PR #42122](https://github.com/BerriAI/litellm/pull/42122) - Map OCR page markdown onto the generation output - [PR #42267](https://github.com/BerriAI/litellm/pull/42267) - Deliver every distinct alert queued in one flush window - [PR #42314](https://github.com/BerriAI/litellm/pull/42314) - Make flush() survive an event loop change - [PR #42355](https://github.com/BerriAI/litellm/pull/42355) - Per-team success and error sampling rates for the Arize AX callback - [PR #42383](https://github.com/BerriAI/litellm/pull/42383) - Price terminal Responses stream events from their inner response - [PR #42385](https://github.com/BerriAI/litellm/pull/42385) - Map completions, images, speech, transcription and moderation output onto the Langfuse generation output - [PR #42394](https://github.com/BerriAI/litellm/pull/42394) - Json.dumps with default=str so non-serializable metadata does not crash batch flush - [PR #42424](https://github.com/BerriAI/litellm/pull/42424) - Record the GenAI exception event through the Logs API on both OpenTelemetry lines - [PR #42431](https://github.com/BerriAI/litellm/pull/42431) - Map rerank and search output and the OCR, image edit and search input onto the Langfuse generation - [PR #42444](https://github.com/BerriAI/litellm/pull/42444) - Keep text completion choice fields beside the synthesized message - [PR #42537](https://github.com/BerriAI/litellm/pull/42537) - Replay approved pre-call guardrail snapshots - [PR #42774](https://github.com/BerriAI/litellm/pull/42774) - Scan the exceeded budget wording linearly so a crafted error message cannot stall the proxy - [PR #42778](https://github.com/BerriAI/litellm/pull/42778) - Pass provider response headers to callbacks on every endpoint - [PR #42824](https://github.com/BerriAI/litellm/pull/42824) - Root post-response service spans in their own trace linked to the request - [PR #42826](https://github.com/BerriAI/litellm/pull/42826) - Add model_group label to deployment request and rate limit metrics - [PR #42966](https://github.com/BerriAI/litellm/pull/42966) - Upload fresh events first, drop terminal failures and hour-old retries by default, opt-in adaptive concurrency - [PR #43022](https://github.com/BerriAI/litellm/pull/43022) ### Secret Managers - Restore secret scheduled for deletion instead of failing CreateSecret - [PR #42454](https://github.com/BerriAI/litellm/pull/42454) - Route secret resolution through native Rust backends - [PR #42619](https://github.com/BerriAI/litellm/pull/42619) ## Spend Tracking, Budgets and Rate Limiting ### Cost tracking - Honor deployment pricing for image generation - [PR #39311](https://github.com/BerriAI/litellm/pull/39311) - Bill batch prompts above 272K at OpenAI's long-context batch tier - [PR #39861](https://github.com/BerriAI/litellm/pull/39861) - Attribute CLI session spend to the per-user cli-session alias instead of the hashed session token - [PR #40541](https://github.com/BerriAI/litellm/pull/40541) - Warn and count $0 cost on billable requests - [PR #42345](https://github.com/BerriAI/litellm/pull/42345) - Honor per-second custom pricing on chat completions for every provider - [PR #42403](https://github.com/BerriAI/litellm/pull/42403) - Return 400 from /spend/calculate for a model with no pricing row - [PR #42497](https://github.com/BerriAI/litellm/pull/42497) - Capture-rate check of LiteLLM spend against the OpenAI bill - [PR #43044](https://github.com/BerriAI/litellm/pull/43044) - Apply a deployment's pricing override to realtime sessions - [PR #43114](https://github.com/BerriAI/litellm/pull/43114) ### Budgets - Renew budget reservation counter TTL while the request is in flight - [PR #40322](https://github.com/BerriAI/litellm/pull/40322) - Enforce virtual key budgets for JEV test routing - [PR #41879](https://github.com/BerriAI/litellm/pull/41879) - Return 422 instead of 429 for BudgetExceededError - [PR #42097](https://github.com/BerriAI/litellm/pull/42097) - Release unclaimed budget reservations at request end - [PR #42304](https://github.com/BerriAI/litellm/pull/42304) - Await budget redis pipeline before sync reads (internal copy of #32618) - [PR #43125](https://github.com/BerriAI/litellm/pull/43125) ### Rate limiting - Enforce tpm_limit and rpm_limit set on tag objects - [PR #41807](https://github.com/BerriAI/litellm/pull/41807) - Share model rate-limit buckets between a model_group_alias and its target - [PR #42516](https://github.com/BerriAI/litellm/pull/42516) ## MCP Gateway ### MCP Gateway - Handle split UTF-8 routing previews - [PR #34919](https://github.com/BerriAI/litellm/pull/34919) - Paginate prompt and resource discovery - [PR #39189](https://github.com/BerriAI/litellm/pull/39189) - Keep oauth scopes in admin api credential redaction - [PR #39805](https://github.com/BerriAI/litellm/pull/39805) - Keep server lists stable across refreshes - [PR #41074](https://github.com/BerriAI/litellm/pull/41074) - Apply post-call rewrites without stale structured output - [PR #41530](https://github.com/BerriAI/litellm/pull/41530) - Tools/call no longer 404s on a worker that has not served tools/list - [PR #42072](https://github.com/BerriAI/litellm/pull/42072) - Explain missing public client dependencies - [PR #42148](https://github.com/BerriAI/litellm/pull/42148) - Keep config-defined servers read-only - [PR #42299](https://github.com/BerriAI/litellm/pull/42299) - Restore legacy SSE and bounded cancellation cleanup - [PR #42382](https://github.com/BerriAI/litellm/pull/42382) - Cap an agent key's tools at what the invoking user and team may call - [PR #42478](https://github.com/BerriAI/litellm/pull/42478) - Preserve discovery attribution and sanitize logging headers - [PR #42541](https://github.com/BerriAI/litellm/pull/42541) - Preserve credential authority in DCR bridge authentication - [PR #42563](https://github.com/BerriAI/litellm/pull/42563) - Reject origins outside the configured allowlist - [PR #42649](https://github.com/BerriAI/litellm/pull/42649) - Return 401 challenge for REST token-exchange tool calls without a subject token - [PR #42782](https://github.com/BerriAI/litellm/pull/42782) - Forward caller bearer on REST oauth_delegate tool calls - [PR #42787](https://github.com/BerriAI/litellm/pull/42787) - Keep tool attribution on guardrail-blocked REST calls - [PR #42790](https://github.com/BerriAI/litellm/pull/42790) - Reject duplicate MCP server names and aliases - [PR #42791](https://github.com/BerriAI/litellm/pull/42791) - Allow ["\*"] wildcard in mcp_tool_permissions to grant all current and future tools - [PR #43108](https://github.com/BerriAI/litellm/pull/43108) ## Performance / Loadbalancing / Reliability improvements ### Auto Router and model routing - Add configurable provider affinity header mapping - [PR #41033](https://github.com/BerriAI/litellm/pull/41033) - Make context-window escalation opt-in - [PR #41872](https://github.com/BerriAI/litellm/pull/41872) - Add JEV classifier alongside LLM classifier - [PR #41886](https://github.com/BerriAI/litellm/pull/41886) - Add NotFoundErrorRetries so a retry policy can pin 404 retries - [PR #42045](https://github.com/BerriAI/litellm/pull/42045) - Skip cooldown for background response cost poll 404s - [PR #42046](https://github.com/BerriAI/litellm/pull/42046) - Stamp model_group when retrieving a batch, so batch tokens are attributable (internal copy of #38499) - [PR #42062](https://github.com/BerriAI/litellm/pull/42062) - Native compact-to-fit across conversation APIs - [PR #42074](https://github.com/BerriAI/litellm/pull/42074) - Keep prompt caching affinity when the breakpoint moves - [PR #42080](https://github.com/BerriAI/litellm/pull/42080) - Configure heuristic v2 success threshold - [PR #42252](https://github.com/BerriAI/litellm/pull/42252) - Walk every entry of a fallback list after a mid-stream failure - [PR #42283](https://github.com/BerriAI/litellm/pull/42283) - Add group-scoped priority routing strategy - [PR #42378](https://github.com/BerriAI/litellm/pull/42378) - Time-windowed team reservation of deployments via model_info.access_windows - [PR #42398](https://github.com/BerriAI/litellm/pull/42398) - Explain fallback outcome in plain words in the raised error - [PR #42509](https://github.com/BerriAI/litellm/pull/42509) - Serve Responses turns from a sibling when the encrypted content origin has no boundary peer - [PR #43015](https://github.com/BerriAI/litellm/pull/43015) - Match provider-prefixed fallback keys for bare model groups served by wildcard deployments - [PR #43062](https://github.com/BerriAI/litellm/pull/43062) - Honor disable_fallbacks on mid-stream fallback - [PR #43111](https://github.com/BerriAI/litellm/pull/43111) ### Caching, database and runtime - Authenticate sync clusters with IAM credential providers - [PR #40204](https://github.com/BerriAI/litellm/pull/40204) - Coordinate v2 migration startup and qualify container recovery - [PR #40932](https://github.com/BerriAI/litellm/pull/40932) - Record aborted outcome when spend-log cleanup is cancelled at shutdown - [PR #41213](https://github.com/BerriAI/litellm/pull/41213) - Wait for the spend-log table before creating startup views - [PR #41974](https://github.com/BerriAI/litellm/pull/41974) - Stop re-sending un-resendable spend batches from the Redis buffer - [PR #41994](https://github.com/BerriAI/litellm/pull/41994) - Park requeued spend logs in Redis so they survive a pod restart during a DB outage - [PR #42022](https://github.com/BerriAI/litellm/pull/42022) - Default to the v2 migration resolver - [PR #42105](https://github.com/BerriAI/litellm/pull/42105) - Publish auth cache invalidations in the background so a wedged coordination Redis cannot stall user updates - [PR #42534](https://github.com/BerriAI/litellm/pull/42534) - Honor DATABASE_DISABLE_PREPARED_STATEMENTS in the litellm CLI - [PR #42556](https://github.com/BerriAI/litellm/pull/42556) - Keep embedding cache hits aligned with request inputs - [PR #42571](https://github.com/BerriAI/litellm/pull/42571) - Keep the in-flight daily spend batch when shutdown cancels the flush - [PR #42593](https://github.com/BerriAI/litellm/pull/42593) - Fail parked DB lookups at a deadline and flip readiness while they stall - [PR #42654](https://github.com/BerriAI/litellm/pull/42654) - Do not requeue a daily spend batch whose commit already left for postgres - [PR #42786](https://github.com/BerriAI/litellm/pull/42786) - Apply user_api_key_cache_max_size to the key object partition - [PR #42796](https://github.com/BerriAI/litellm/pull/42796) - Stamp provider on sync cache-hit logs so responses spend logs record provider - [PR #42830](https://github.com/BerriAI/litellm/pull/42830) ### Native Rust runtime (opt-in) - Preserve Python defaults with opt-in Rust dispatch - [PR #42174](https://github.com/BerriAI/litellm/pull/42174) - Add native disk cache backend - [PR #42311](https://github.com/BerriAI/litellm/pull/42311) - Add native S3 cache backend - [PR #42313](https://github.com/BerriAI/litellm/pull/42313) - Add native Valkey semantic cache backend - [PR #42316](https://github.com/BerriAI/litellm/pull/42316) - Serve RedisClusterCache natively as a Redis topology - [PR #42317](https://github.com/BerriAI/litellm/pull/42317) - Serve Redis Semantic caches natively in Rust - [PR #42319](https://github.com/BerriAI/litellm/pull/42319) - Native Azure Blob response cache backend - [PR #42321](https://github.com/BerriAI/litellm/pull/42321) - Serve QdrantSemanticCache natively from Rust - [PR #42324](https://github.com/BerriAI/litellm/pull/42324) - Add native GCS object-store cache backend - [PR #42325](https://github.com/BerriAI/litellm/pull/42325) - Align the cache crates with Python and wire every native backend - [PR #42530](https://github.com/BerriAI/litellm/pull/42530) - Dispatch Python logging through the Rust diagnostics processor - [PR #42616](https://github.com/BerriAI/litellm/pull/42616) - Stamp x-litellm-rust on native sync and async streams at the bridge boundary - [PR #42758](https://github.com/BerriAI/litellm/pull/42758) - Shape Anthropic Messages requests natively - [PR #42982](https://github.com/BerriAI/litellm/pull/42982) - Read secrets through Python from Rust routes and declare Rust-only routes with NO_PYTHON - [PR #43057](https://github.com/BerriAI/litellm/pull/43057) - Hand upstream response headers to the native Messages stream - [PR #43178](https://github.com/BerriAI/litellm/pull/43178) - Traverse and release the retained headers dict - [PR #43274](https://github.com/BerriAI/litellm/pull/43274) ## Documentation Updates ### Documentation - Stop advertising sk-1234 as the master key in shipped configs and examples - [PR #42011](https://github.com/BerriAI/litellm/pull/42011) ## Tests, CI and Internal Changes These 184 PRs change tests, CI, contributor tooling, release packaging, or Rust runtime scaffolding that is not yet wired into a user-facing path. They do not change proxy or SDK behavior on their own
Tests (108) - Replace custom endpoints_client with provider SDK clients - [PR #34358](https://github.com/BerriAI/litellm/pull/34358) - Assert cache-priced vertex grok rows advertise supports_prompt_caching - [PR #41526](https://github.com/BerriAI/litellm/pull/41526) - Cover actor edges and wildcard models - [PR #41769](https://github.com/BerriAI/litellm/pull/41769) - Cover dashboard form journeys - [PR #41773](https://github.com/BerriAI/litellm/pull/41773) - Cover chat and responses registry gaps - [PR #41794](https://github.com/BerriAI/litellm/pull/41794) - Cover legacy lowest TPM selection - [PR #41795](https://github.com/BerriAI/litellm/pull/41795) - Endpoint, breakdown component and failure support in the cost harness - [PR #41999](https://github.com/BerriAI/litellm/pull/41999) - Embeddings, rerank, completions and moderations cost cases - [PR #42020](https://github.com/BerriAI/litellm/pull/42020) - Audio, image and per-unit cost cases - [PR #42024](https://github.com/BerriAI/litellm/pull/42024) - Passthrough route cost cases - [PR #42028](https://github.com/BerriAI/litellm/pull/42028) - Pricing dimension and provider reported cost cases - [PR #42035](https://github.com/BerriAI/litellm/pull/42035) - Restore scoped execution and credential isolation regressions - [PR #42050](https://github.com/BerriAI/litellm/pull/42050) - Restore MCP OAuth happy-path coverage (LIT-3467) - [PR #42051](https://github.com/BerriAI/litellm/pull/42051) - Provider wire cost cases - [PR #42052](https://github.com/BerriAI/litellm/pull/42052) - Proxy behaviour cost cases - [PR #42060](https://github.com/BerriAI/litellm/pull/42060) - Add autorouter estimate keys to the GCS pub/sub spend-log golden - [PR #42061](https://github.com/BerriAI/litellm/pull/42061) - Batch and realtime cost cases - [PR #42066](https://github.com/BerriAI/litellm/pull/42066) - Migrate phase 6 provider unit tests to tests/unit - [PR #42107](https://github.com/BerriAI/litellm/pull/42107) - Migrate wave 1 phase 2 legacy unit tests to tests/unit - [PR #42108](https://github.com/BerriAI/litellm/pull/42108) - Migrate phase 5 provider unit tests to tests/unit - [PR #42109](https://github.com/BerriAI/litellm/pull/42109) - Migrate bedrock, baseten and base_llm batch tests to tests/unit - [PR #42110](https://github.com/BerriAI/litellm/pull/42110) - Migrate wave 1 phase 8 legacy llm tests to tests/unit - [PR #42112](https://github.com/BerriAI/litellm/pull/42112) - Block external sockets at import time and add a socket policy regression test - [PR #42113](https://github.com/BerriAI/litellm/pull/42113) - Migrate phase 7 provider unit tests to tests/unit - [PR #42114](https://github.com/BerriAI/litellm/pull/42114) - Migrate nvidia, oci, ocr, oobabooga and openai legacy tests to tests/unit - [PR #42115](https://github.com/BerriAI/litellm/pull/42115) - Migrate phase 9 legacy llm provider tests to tests/unit - [PR #42117](https://github.com/BerriAI/litellm/pull/42117) - Migrate wave 1 phase 3 anthropic, apiserpent, azure and azure_ai legacy tests - [PR #42118](https://github.com/BerriAI/litellm/pull/42118) - Boot the unified Google proxy fixture with a real master key - [PR #42120](https://github.com/BerriAI/litellm/pull/42120) - Migrate wave 1 phase 1 legacy tests to tests/unit - [PR #42123](https://github.com/BerriAI/litellm/pull/42123) - Deflake fuzzy picker, breached-password HIBP, and MCP stdio timeout tests (rolling deflake 2026-09-22) - [PR #42125](https://github.com/BerriAI/litellm/pull/42125) - Migrate openai, openai_like and openrouter legacy tests to tests/unit - [PR #42128](https://github.com/BerriAI/litellm/pull/42128) - Migrate phase 16 legacy tests to tests/unit - [PR #42131](https://github.com/BerriAI/litellm/pull/42131) - Migrate phase 14 wave 2 provider tests to tests/unit - [PR #42132](https://github.com/BerriAI/litellm/pull/42132) - Make every tests/unit directory a package so pytest collection is unique - [PR #42135](https://github.com/BerriAI/litellm/pull/42135) - Migrate phase 15 legacy tests to tests/unit - [PR #42136](https://github.com/BerriAI/litellm/pull/42136) - Migrate phase 12 legacy llm provider tests to tests/unit - [PR #42137](https://github.com/BerriAI/litellm/pull/42137) - Migrate legacy provider tests to tests/unit (wave 2, phase 13) - [PR #42145](https://github.com/BerriAI/litellm/pull/42145) - Migrate a2a_protocol legacy tests to tests/unit (wave 3 phase 17) - [PR #42160](https://github.com/BerriAI/litellm/pull/42160) - Cover the release-to-release upgrade path - [PR #42294](https://github.com/BerriAI/litellm/pull/42294) - Fix stale budget-status and bad-database-url assertions - [PR #42339](https://github.com/BerriAI/litellm/pull/42339) - Add conversational matrix across chat, messages and responses - [PR #42359](https://github.com/BerriAI/litellm/pull/42359) - Isolate the agent read-through singleton between unknown-agent tests - [PR #42389](https://github.com/BerriAI/litellm/pull/42389) - Move Xiaomi MiMo coverage from live e2e to the providers wire shard - [PR #42395](https://github.com/BerriAI/litellm/pull/42395) - Chain a proxy-issued previous_response_id in the cost suite - [PR #42396](https://github.com/BerriAI/litellm/pull/42396) - Check the MCP Tools tab against the upstream's own tools/list - [PR #42397](https://github.com/BerriAI/litellm/pull/42397) - Make bedrock collector and secret scan timing tests deterministic - [PR #42405](https://github.com/BerriAI/litellm/pull/42405) - Assert a cooldown reaches a sibling replica within the 1s Redis read interval - [PR #42422](https://github.com/BerriAI/litellm/pull/42422) - Make the detailed-timing receive-anchor test timezone independent - [PR #42429](https://github.com/BerriAI/litellm/pull/42429) - Give every ui settings endpoint test a fresh settings store - [PR #42430](https://github.com/BerriAI/litellm/pull/42430) - Add secret manager lanes for HashiCorp Vault and CyberArk Conjur - [PR #42503](https://github.com/BerriAI/litellm/pull/42503) - Run the memory cell alone on the shared stack - [PR #42518](https://github.com/BerriAI/litellm/pull/42518) - Move unified_google_tests to gemini-3.5-flash-lite - [PR #42520](https://github.com/BerriAI/litellm/pull/42520) - Drop route dispatch assertions, test the bridge directly - [PR #42536](https://github.com/BerriAI/litellm/pull/42536) - One request lands the same spend on every surface - [PR #42540](https://github.com/BerriAI/litellm/pull/42540) - Guard leaked cassette patches and make injected-transport embedding tests immune - [PR #42542](https://github.com/BerriAI/litellm/pull/42542) - Hold every worker under an idle RSS budget before any traffic - [PR #42552](https://github.com/BerriAI/litellm/pull/42552) - Count a zombie grandchild as gone in the migrate deploy timeout test - [PR #42570](https://github.com/BerriAI/litellm/pull/42570) - Make two proxy-infra tests independent of sibling-test state - [PR #42581](https://github.com/BerriAI/litellm/pull/42581) - Point unit tests at model ids still in the cost map - [PR #42606](https://github.com/BerriAI/litellm/pull/42606) - Accept the per-size image cost keys in the price-map schema check - [PR #42612](https://github.com/BerriAI/litellm/pull/42612) - Point image-generation deployment price test at a live gemini row - [PR #42615](https://github.com/BerriAI/litellm/pull/42615) - Point CircleCI-only suites at models still in the cost map - [PR #42617](https://github.com/BerriAI/litellm/pull/42617) - Regression tests for August provider translation and streaming bugs - [PR #42621](https://github.com/BerriAI/litellm/pull/42621) - Regression tests for August cost tracking and budgeting bugs - [PR #42622](https://github.com/BerriAI/litellm/pull/42622) - Drop legacy InvalidStatusCode tests and pin websockets imports - [PR #42624](https://github.com/BerriAI/litellm/pull/42624) - Tolerate provider-side flakes on five full-suite cells - [PR #42628](https://github.com/BerriAI/litellm/pull/42628) - Repoint the Azure image cost test at gpt-image-2 - [PR #42631](https://github.com/BerriAI/litellm/pull/42631) - Raise the post-success hook error from a guardrail in the failure-hook regression - [PR #42646](https://github.com/BerriAI/litellm/pull/42646) - Allow skipped nodes and drop the shard cap - [PR #42687](https://github.com/BerriAI/litellm/pull/42687) - Add read-replica routing harness to the CircleCI integration suite - [PR #42692](https://github.com/BerriAI/litellm/pull/42692) - Regression tests for July provider translation, routing and streaming bugs - [PR #42693](https://github.com/BerriAI/litellm/pull/42693) - Regression tests for July cost tracking, budgeting and spend bugs - [PR #42694](https://github.com/BerriAI/litellm/pull/42694) - Add MCP gateway coverage wave 1 with a dedicated mcp shard and proxy coverage artifact - [PR #42711](https://github.com/BerriAI/litellm/pull/42711) - Model the blocking OCR hook as a guardrail so its raise propagates - [PR #42775](https://github.com/BerriAI/litellm/pull/42775) - Deterministic integration audit of the v3 platform relay - [PR #42781](https://github.com/BerriAI/litellm/pull/42781) - Cover customer reported cache key, cache_control, bedrock request id, responses schema, scim and tag budget contracts - [PR #42785](https://github.com/BerriAI/litellm/pull/42785) - Assert /v1/responses usage reports Anthropic system cache write then read - [PR #42855](https://github.com/BerriAI/litellm/pull/42855) - Assert /v1/models reports max_input_tokens and max_output_tokens - [PR #42858](https://github.com/BerriAI/litellm/pull/42858) - Cover per key tag rpm limits and tag budget_duration resets - [PR #42859](https://github.com/BerriAI/litellm/pull/42859) - Edge-case matrices for malformed token limits and callback_settings shapes - [PR #42895](https://github.com/BerriAI/litellm/pull/42895) - Take keys out of the legacy proxy, enterprise and mcp unit tests before moving them - [PR #42901](https://github.com/BerriAI/litellm/pull/42901) - Assert the images sent to Ollama instead of echoing them through response - [PR #42905](https://github.com/BerriAI/litellm/pull/42905) - Stop a comprehension variable from shadowing the body() helper - [PR #42906](https://github.com/BerriAI/litellm/pull/42906) - Run the files peak-memory guards without coverage tracing - [PR #42914](https://github.com/BerriAI/litellm/pull/42914) - Patch create_mcp_server_if_identifier_free in the store-model-in-db MCP tests - [PR #42916](https://github.com/BerriAI/litellm/pull/42916) - Run the cache-hit redis outage test on the shared owned_redis helper - [PR #42925](https://github.com/BerriAI/litellm/pull/42925) - Give the logout specs their own admin session - [PR #42930](https://github.com/BerriAI/litellm/pull/42930) - Accept the otel cost write as a linked root trace - [PR #42931](https://github.com/BerriAI/litellm/pull/42931) - Hide the LiteAdmin button in the shared admin session - [PR #43033](https://github.com/BerriAI/litellm/pull/43033) - Pin end-user and tag attribution from Codex-style headers on /v1/responses - [PR #43093](https://github.com/BerriAI/litellm/pull/43093) - Cover Azure code_interpreter container files by native id with a service-account key - [PR #43122](https://github.com/BerriAI/litellm/pull/43122) - Reorganize core crate tests and split cache and OCR suites - [PR #43177](https://github.com/BerriAI/litellm/pull/43177) - Agent clients for Claude Code, Codex and opencode - [PR #43181](https://github.com/BerriAI/litellm/pull/43181) - Move tests/test_litellm root and small trees into tests/unit - [PR #43186](https://github.com/BerriAI/litellm/pull/43186) - Move tests/test_litellm/llms into tests/unit/llms - [PR #43191](https://github.com/BerriAI/litellm/pull/43191) - Move tests/test_litellm integrations and secret_managers into tests/unit - [PR #43194](https://github.com/BerriAI/litellm/pull/43194) - Move tests/test_litellm core utils, routing, responses, caching and rust_bridge into tests/unit - [PR #43199](https://github.com/BerriAI/litellm/pull/43199) - Drop the repeated UNIT_FLAG key in test_unit_shard_missing_paths - [PR #43212](https://github.com/BerriAI/litellm/pull/43212) - Stop CI tests from downloading tokenizer files and images - [PR #43257](https://github.com/BerriAI/litellm/pull/43257) - Fix stale and state-leaking tests red on scheduled CircleCI - [PR #43266](https://github.com/BerriAI/litellm/pull/43266) - Finish the non-proxy half of tests/test_litellm - [PR #43281](https://github.com/BerriAI/litellm/pull/43281) - Port langfuse callbacks-in-db coverage to the local harness - [PR #43282](https://github.com/BerriAI/litellm/pull/43282) - Run the Langfuse DB-callback test on its own scratch database - [PR #43288](https://github.com/BerriAI/litellm/pull/43288) - Scope the management proxy fixture to its package so its spend monitor cannot race the spend tests - [PR #43302](https://github.com/BerriAI/litellm/pull/43302) - Assert only litellm-owned batch behavior and move the blank S3 env pin to an integration test - [PR #43321](https://github.com/BerriAI/litellm/pull/43321) - Report batch cleanup leftovers as a plain UserWarning on rc/1.104.0 - [PR #43406](https://github.com/BerriAI/litellm/pull/43406) - Skip the LangSmith batch serialization e2e on rc/1.104.0 until CI has a LangSmith key - [PR #43418](https://github.com/BerriAI/litellm/pull/43418) - Stop test modules from putting their own directory on sys.path on rc/1.104.0 - [PR #43420](https://github.com/BerriAI/litellm/pull/43420)
CI (27) - Wire tests/unit into CircleCI and keep draining GHA shards green - [PR #42103](https://github.com/BerriAI/litellm/pull/42103) - Excuse retired test-quality rules in the budget ratchet - [PR #42116](https://github.com/BerriAI/litellm/pull/42116) - Fix the stage-mirror batch reds and keep a redacted pytest log - [PR #42143](https://github.com/BerriAI/litellm/pull/42143) - Let the install smoke test boot its key-less proxy config - [PR #42296](https://github.com/BerriAI/litellm/pull/42296) - Skip cost map file checks on PRs that leave the cost map untouched - [PR #42406](https://github.com/BerriAI/litellm/pull/42406) - Print add-mask lines only under GitHub Actions - [PR #42423](https://github.com/BerriAI/litellm/pull/42423) - Allowlist _render_json in the recursive detector - [PR #42442](https://github.com/BerriAI/litellm/pull/42442) - Route credential, cost map, and UI login calls to the control plane - [PR #42506](https://github.com/BerriAI/litellm/pull/42506) - Drop dead misc shard paths and skip missing paths with a warning - [PR #42603](https://github.com/BerriAI/litellm/pull/42603) - Move the compat-matrix populator from a GCE VM to a Render cron job - [PR #42608](https://github.com/BerriAI/litellm/pull/42608) - Close open pull requests superseded by a merged fix on their linked issue - [PR #42609](https://github.com/BerriAI/litellm/pull/42609) - Remove the unused create-release workflow - [PR #42696](https://github.com/BerriAI/litellm/pull/42696) - Add merge smoke checks workflow with loopback-only harness and 11 curated cases - [PR #42709](https://github.com/BerriAI/litellm/pull/42709) - Drop litellm_internal_staging and litellm_oss_staging references, main is the only trunk - [PR #42745](https://github.com/BerriAI/litellm/pull/42745) - Add tests-only CircleCI pipeline with coverage and docs validation - [PR #42773](https://github.com/BerriAI/litellm/pull/42773) - Fix the litellm-tests unit job (sysmon, codecov on failure, env -i allowlist, selection errors, reruns param) - [PR #42900](https://github.com/BerriAI/litellm/pull/42900) - Move caching, proxy-extras, gateway and enterprise tests into tests/unit and run them from litellm-tests - [PR #42902](https://github.com/BerriAI/litellm/pull/42902) - Move tests/proxy_unit_tests to tests/unit/proxy and run the proxy-db shards from litellm-tests - [PR #42903](https://github.com/BerriAI/litellm/pull/42903) - Move provider-independent MCP tests into tests/unit and run mcp-integration from litellm-tests - [PR #42904](https://github.com/BerriAI/litellm/pull/42904) - Resolve and install the Claude Code CLI per run - [PR #43038](https://github.com/BerriAI/litellm/pull/43038) - Skip unpublished npm versions in the Claude Code PR-gate resolver - [PR #43053](https://github.com/BerriAI/litellm/pull/43053) - Run the claude_code harness unit-test trees in the lint job - [PR #43077](https://github.com/BerriAI/litellm/pull/43077) - Cut rc/<X.Y.0> off main every Friday at 3am Pacific - [PR #43121](https://github.com/BerriAI/litellm/pull/43121) - Run migrated unit selections on every event in legacy GHA shards - [PR #43182](https://github.com/BerriAI/litellm/pull/43182) - Drop main and litellm_\* branch filters from the CircleCI litellm-main workflows - [PR #43272](https://github.com/BerriAI/litellm/pull/43272) - Stop stale CI reds, keep unit tests off the host env, retry CyberArk policy conflicts - [PR #43294](https://github.com/BerriAI/litellm/pull/43294) - Cut CircleCI wall time without loosening test isolation - [PR #43347](https://github.com/BerriAI/litellm/pull/43347)
Code quality and contributor tooling (16) - Remove 1,173 Any errors across 169 backend files - [PR #40251](https://github.com/BerriAI/litellm/pull/40251) - Define the tier contract for unit, integration and e2e - [PR #42099](https://github.com/BerriAI/litellm/pull/42099) - Replace Any with proven types in 30 files - [PR #42127](https://github.com/BerriAI/litellm/pull/42127) - Replace Any with proven types in 32 files - [PR #42220](https://github.com/BerriAI/litellm/pull/42220) - Extract explicit operation context and dispatch - [PR #42292](https://github.com/BerriAI/litellm/pull/42292) - Cap comprehensions at one for and one if clause (LIT014) - [PR #42650](https://github.com/BerriAI/litellm/pull/42650) - Clear fresh tech debt from the last 24 hours (rolling, 2026-09-06 to 2026-09-24) - [PR #42710](https://github.com/BerriAI/litellm/pull/42710) - Replace Any with proven types in 5 files - [PR #42722](https://github.com/BerriAI/litellm/pull/42722) - Add LIT013 flagging \*-ok suppressions that suppress nothing and remove the 240 stale ones - [PR #42793](https://github.com/BerriAI/litellm/pull/42793) - Drop empty sections from the PR body and tighten the User Flow - [PR #42794](https://github.com/BerriAI/litellm/pull/42794) - Simplify pull request template into plain English questions - [PR #42813](https://github.com/BerriAI/litellm/pull/42813) - Restore the full pull request template (reverts #42813) - [PR #42828](https://github.com/BerriAI/litellm/pull/42828) - Declare litellm-owned kwargs as typed objects and derive the lists from their fields - [PR #42843](https://github.com/BerriAI/litellm/pull/42843) - Replace Any with proven types in 13 files - [PR #42937](https://github.com/BerriAI/litellm/pull/42937) - Require UI before/after screenshots and intentional UX change note in PR template - [PR #43021](https://github.com/BerriAI/litellm/pull/43021) - Carve harness tests out of the no-unit-tests hard rule - [PR #43076](https://github.com/BerriAI/litellm/pull/43076)
Rust runtime internals (29) - Split token counter backends - [PR #42165](https://github.com/BerriAI/litellm/pull/42165) - Add typed secret managers and shared auth adapters - [PR #42173](https://github.com/BerriAI/litellm/pull/42173) - Scaffold cache foundation for Python parity - [PR #42196](https://github.com/BerriAI/litellm/pull/42196) - Preserve Python settings semantics at the native boundary - [PR #42300](https://github.com/BerriAI/litellm/pull/42300) - Add CyberArk Conjur secret manager backend - [PR #42303](https://github.com/BerriAI/litellm/pull/42303) - Add HashiCorp Vault secret manager crate - [PR #42308](https://github.com/BerriAI/litellm/pull/42308) - Add Azure Key Vault secret manager backend - [PR #42309](https://github.com/BerriAI/litellm/pull/42309) - Add cache and secret migration foundations - [PR #42328](https://github.com/BerriAI/litellm/pull/42328) - Declare _CacheTestHandle.valkey_semantic in native stub - [PR #42364](https://github.com/BerriAI/litellm/pull/42364) - Keep native Redis semantic binding and Qdrant batch writes after merge - [PR #42379](https://github.com/BerriAI/litellm/pull/42379) - Align secret manager operation contexts - [PR #42480](https://github.com/BerriAI/litellm/pull/42480) - Add python-compat crate for Python data formats - [PR #42510](https://github.com/BerriAI/litellm/pull/42510) - Add standalone cost calculator - [PR #42604](https://github.com/BerriAI/litellm/pull/42604) - Add immutable model catalog crate - [PR #42605](https://github.com/BerriAI/litellm/pull/42605) - Add a guarded native response-cache resolver foundation - [PR #42769](https://github.com/BerriAI/litellm/pull/42769) - Add native dispatch foundation - [PR #42799](https://github.com/BerriAI/litellm/pull/42799) - Extend native dispatch foundation to chat completions, responses, and messages - [PR #42805](https://github.com/BerriAI/litellm/pull/42805) - Declare above_32k cost fields on ModelInfo - [PR #42856](https://github.com/BerriAI/litellm/pull/42856) - Move tests.rs files inline or under tests/ and drop autotests = false - [PR #43028](https://github.com/BerriAI/litellm/pull/43028) - Extract the host coroutine into its own crate - [PR #43129](https://github.com/BerriAI/litellm/pull/43129) - Add Rust registry validation - [PR #43136](https://github.com/BerriAI/litellm/pull/43136) - Replace Framer trait with tokio-util codecs - [PR #43193](https://github.com/BerriAI/litellm/pull/43193) - Hand out an owned Client and route all providers through the pool - [PR #43245](https://github.com/BerriAI/litellm/pull/43245) - Move credential inheritance and SDK limits into a driver preflight - [PR #43259](https://github.com/BerriAI/litellm/pull/43259) - Preserve nested optional import failures - [PR #43265](https://github.com/BerriAI/litellm/pull/43265) - Promote anthropic messages out of experimental_pass_through - [PR #43269](https://github.com/BerriAI/litellm/pull/43269) - Prepare inference and auth foundations for the gateway - [PR #43287](https://github.com/BerriAI/litellm/pull/43287) - Add config, router and gateway crates - [PR #43289](https://github.com/BerriAI/litellm/pull/43289) - Expand logging and test coverage across gateway and Anthropic messages - [PR #43295](https://github.com/BerriAI/litellm/pull/43295)
Release and packaging (4) - Bump litellm-enterprise 0.1.69 -> 0.1.70, litellm-proxy-extras 0.4.100 -> 0.4.101, litellm 1.103.0 -> 1.104.0 - [PR #42633](https://github.com/BerriAI/litellm/pull/42633) - Bump litellm-enterprise 0.1.70 -> 0.1.71, litellm-proxy-extras 0.4.101 -> 0.4.102 - [PR #43120](https://github.com/BerriAI/litellm/pull/43120) - Rebuild Admin UI bundle for rc/1.104.0 - [PR #43372](https://github.com/BerriAI/litellm/pull/43372) - Remove the top-N key cap, its follow-ups and the daily global spend rollup code from rc/1.104.0 - [PR #43385](https://github.com/BerriAI/litellm/pull/43385)
### PR roll-up by ownership area Customer-facing PRs: **447**. Tests, CI and internal PRs: **184**. Total: **631** - Models & Providers: 236 - Other (tests, CI, internal): 184 - Performance: 47 - Auth & Management: 43 - LLM API Endpoints: 25 - Logging: 24 - UI: 24 - MCP: 18 - Spend / Budgets / Rate Limits: 15 - Guardrails: 12 - Secret Managers: 2 - Docs: 1 ## New Contributors - [@ahamedshaik16](https://github.com/ahamedshaik16) made their first contribution in [PR #42966](https://github.com/BerriAI/litellm/pull/42966) - [@chopratejas](https://github.com/chopratejas) made their first contribution in [PR #41560](https://github.com/BerriAI/litellm/pull/41560) - [@doramirdor](https://github.com/doramirdor) made their first contribution in [PR #33227](https://github.com/BerriAI/litellm/pull/33227) - [@kerry-berri](https://github.com/kerry-berri) made their first contribution in [PR #40429](https://github.com/BerriAI/litellm/pull/40429) - [@kumarpriyanshu09](https://github.com/kumarpriyanshu09) made their first contribution in [PR #41526](https://github.com/BerriAI/litellm/pull/41526) - [@patel-26meet](https://github.com/patel-26meet) made their first contribution in [PR #39311](https://github.com/BerriAI/litellm/pull/39311) - [@Pawan-Shahane](https://github.com/Pawan-Shahane) made their first contribution in [PR #41979](https://github.com/BerriAI/litellm/pull/41979) - [@philschmid](https://github.com/philschmid) made their first contribution in [PR #42465](https://github.com/BerriAI/litellm/pull/42465) - [@PhimmStraiker](https://github.com/PhimmStraiker) made their first contribution in [PR #41880](https://github.com/BerriAI/litellm/pull/41880) - [@SiluPanda](https://github.com/SiluPanda) made their first contribution in [PR #40204](https://github.com/BerriAI/litellm/pull/40204) - [@togear](https://github.com/togear) made their first contribution in [PR #41033](https://github.com/BerriAI/litellm/pull/41033) - [@Zechereh](https://github.com/Zechereh) made their first contribution in [PR #41099](https://github.com/BerriAI/litellm/pull/41099) ## Full Changelog [Compare release contents on GitHub](https://github.com/BerriAI/litellm/compare/v1.103.0-rc.1...v1.104.0-rc.1) ## Related pages - [Overview](https://docs.litellm.ai/release_notes/index.md) - [v1.104.0rc2](https://docs.litellm.ai/release_notes/v1.104.0rc2/v1-104-0-rc-2.md)