--- title: "v1.83.14 - GPT-5.5, Prompt Compression & Memory API" url: "/release_notes/v1.83.14/v1-83-14" canonical_url: "https://docs.litellm.ai/release_notes/v1.83.14/v1-83-14" type: "release-notes" last_updated: "2026-10-09" related: - "/release_notes/v1.84.0/v1-84-0" - "/release_notes/v1.83.10/v1-83-10" --- # v1.83.14 - GPT-5.5, Prompt Compression & Memory API > Index of all LiteLLM docs: https://docs.litellm.ai/llms.txt ## Deploy this version **Docker** ```bash docker run \ -e STORE_MODEL_IN_DB=True \ -p 4000:4000 \ docker.litellm.ai/berriai/litellm:main-v1.83.14-stable ``` **Pip** ```bash pip install litellm==1.83.14 ``` ## Key Highlights - **Day-0 GPT-5.5 and GPT-5.5 Pro support** — OpenAI and Azure variants ship with full pricing maps, dated snapshots, and Responses-mode routing for the Pro tier. - **Server-side Prompt Compression** — first-class proxy callback that transparently compresses long-context inputs (Claude Code, RAG, document workloads) before they hit the upstream model, with no client opt-in required. - **`/v1/memory` CRUD endpoints** — proxy now exposes a memory store API with Prisma-backed metadata, consumed by the new agent loop. - **LLM-as-a-Judge guardrail** — model-graded post-call guardrail with configurable rubrics, joining the Bedrock / Lakera / Presidio / Noma family. - **MCP OAuth hardening** — discoverable + BYOK authorize/token endpoints are tightened, temporary OAuth sessions are now shared across proxy instances via Redis, and per-server access policy is uniformly enforced across the proxy and broker. - **Per-member team budgets land in production** — individual member budgets, per-member cycle surfacing in the Teams UI, and atomic counter alignment for user/org spend checks. - **Adaptive routing** — opt-in router policy that weights deployments by recent latency/error history on top of the existing wildcard fallback. --- ## New Models / Updated Models #### New Model Support (22 new models) | Provider | Model | Context Window | Input ($/1M tokens) | Output ($/1M tokens) | Mode | | ------------ | -------------------------------------------------------------------------------------- | -------------- | ------------------- | -------------------- | ---------------- | | OpenAI | `gpt-5.5`, `gpt-5.5-2026-04-23` | 1,050,000 | $5.00 | $30.00 | chat | | OpenAI | `gpt-5.5-pro`, `gpt-5.5-pro-2026-04-23` | 1,050,000 | $60.00 | $360.00 | responses | | OpenAI | `gpt-5.4-mini-2026-03-17` | 272,000 | $0.75 | $4.50 | chat | | OpenAI | `gpt-5.4-nano-2026-03-17` | 272,000 | $0.20 | $1.25 | chat | | Azure OpenAI | `azure/gpt-5.5`, `azure/gpt-5.5-2026-04-23` | 1,050,000 | $5.00 | $30.00 | chat | | Azure OpenAI | `azure/gpt-5.5-pro`, `azure/gpt-5.5-pro-2026-04-23` | 1,050,000 | $60.00 | $360.00 | responses | | Azure OpenAI | `azure/gpt-5.4-mini-2026-03-17` | 1,050,000 | $0.75 | $4.50 | chat | | Azure OpenAI | `azure/gpt-5.4-nano-2026-03-17` | 1,050,000 | $0.20 | $1.25 | chat | | AWS Bedrock | `anthropic.claude-mythos-preview` | 1,000,000 | - | - | chat | | AWS Bedrock | `bedrock/us-east-1/zai.glm-5`, `bedrock/us-west-2/zai.glm-5` | 200,000 | $1.00 | $3.20 | chat | | AWS Bedrock | `bedrock/us-east-1/minimax.minimax-m2.5`, `bedrock/us-west-2/minimax.minimax-m2.5` | - | - | - | chat | | Moonshot | `moonshot/kimi-k2.6` | 262,144 | $0.95 | $4.00 | chat | | OpenRouter | `openrouter/anthropic/claude-opus-4.7` | 1,000,000 | $5.00 | $25.00 | chat | | Gemini | `gemini/gemini-embedding-2`, `gemini-embedding-2`, `vertex_ai/gemini-embedding-2` | 8,192 | $0.20 | - | embedding | | DashScope | `dashscope/qwen-image-2.0`, `dashscope/qwen-image-2.0-pro` | - | - | - | image_generation | #### Features - **[Bedrock](../../docs/providers/bedrock)** - Add GLM-5 and Minimax M2.5 entries with regional aliases - [PR #24423](https://github.com/BerriAI/litellm/pull/24423) - Day-0 support for Claude Mythos Preview via the `bedrock-mantle` endpoint - [PR #26196](https://github.com/BerriAI/litellm/pull/26196) - Allowlist Bedrock Invoke body fields and filter all `anthropic-beta` values - [PR #26148](https://github.com/BerriAI/litellm/pull/26148) - **[OpenAI](../../docs/providers/openai)** - Versioned GPT-5.4 mini / nano snapshots - [PR #26115](https://github.com/BerriAI/litellm/pull/26115) - Add `gpt-5.5` and `gpt-5.5-pro` to the model cost map - [PR #26345](https://github.com/BerriAI/litellm/pull/26345), [PR #26348](https://github.com/BerriAI/litellm/pull/26348) - Day-0 support for GPT-5.5 and GPT-5.5 Pro - [PR #26449](https://github.com/BerriAI/litellm/pull/26449) - **[Azure OpenAI](../../docs/providers/azure)** - `azure/gpt-5.5` + `azure/gpt-5.5-pro` entries with dated variants - [PR #26361](https://github.com/BerriAI/litellm/pull/26361) - **[Gemini](../../docs/providers/gemini)** - Gemini Embedding 2 GA: cost map, blog, and tests - [PR #26391](https://github.com/BerriAI/litellm/pull/26391) - Expand `VideoMetadata` support to all Gemini models - [PR #25767](https://github.com/BerriAI/litellm/pull/25767) - **[Vertex AI](../../docs/providers/vertex)** - Multi-region Vertex hosts (`aiplatform.*.rep.googleapis.com`) - [PR #26281](https://github.com/BerriAI/litellm/pull/26281) - **[DashScope](../../docs/providers/dashscope)** - Image generation support for `qwen-image-2.0` and `qwen-image-2.0-pro` - [PR #25672](https://github.com/BerriAI/litellm/pull/25672) - **[Moonshot](../../docs/providers/moonshot)** - Add `moonshot/kimi-k2.6` to the model registry - [PR #26203](https://github.com/BerriAI/litellm/pull/26203) - **[Anthropic](../../docs/providers/anthropic)** - Migrate retired `claude-3-haiku-20240307` references to `claude-haiku-4-5-20251001` - [PR #26139](https://github.com/BerriAI/litellm/pull/26139) - **General** - Migrate 38 models from legacy `max_tokens` to `max_input_tokens` / `max_output_tokens` - [PR #24422](https://github.com/BerriAI/litellm/pull/24422) ### Bug Fixes - **[Anthropic](../../docs/providers/anthropic)** - Preserve `tool_use` input args in adapter streaming - [PR #24355](https://github.com/BerriAI/litellm/pull/24355) - Strip Gemini thought suffix from streaming `tool_use` id - [PR #25935](https://github.com/BerriAI/litellm/pull/25935) - Skip non-OpenAI file content blocks in file-id discovery helpers - [PR #26228](https://github.com/BerriAI/litellm/pull/26228) - Handle `tool_choice` type `'none'` in messages API - [PR #24457](https://github.com/BerriAI/litellm/pull/24457) - **[Azure](../../docs/providers/azure)** - Preserve `role='assistant'` in streaming with `include_usage` - [PR #24354](https://github.com/BerriAI/litellm/pull/24354) - **[Bedrock](../../docs/providers/bedrock)** - Sort assistant content blocks so text precedes `toolUse` - [PR #24368](https://github.com/BerriAI/litellm/pull/24368) - Above-200k token pricing fix for Claude Sonnet/Opus 4.6 + Sonnet 4.6 `max_input_tokens` to 1M - [PR #24164](https://github.com/BerriAI/litellm/pull/24164) - **[Gemini](../../docs/providers/gemini)** - Filter params from embedding requests - [PR #24370](https://github.com/BerriAI/litellm/pull/24370) - Read web search cost from `model_info` instead of hardcoding - [PR #24372](https://github.com/BerriAI/litellm/pull/24372) - Include DOCUMENT modality tokens in cost calculation - [PR #24410](https://github.com/BerriAI/litellm/pull/24410) - **[Vertex AI](../../docs/providers/vertex)** - Forward `dimensions` parameter in `multimodalembedding` requests - [PR #24415](https://github.com/BerriAI/litellm/pull/24415) - **[Zhipu / GLM](/docs/providers/zai)** - Map non-standard `finish_reason` values - [PR #24373](https://github.com/BerriAI/litellm/pull/24373) - **[OVHcloud](../../docs/providers/ovhcloud)** - Fix tool calling not working - [PR #25948](https://github.com/BerriAI/litellm/pull/25948) - **[Scaleway](../../docs/providers/scaleway)** - Add audio support - [PR #26110](https://github.com/BerriAI/litellm/pull/26110) --- ## LLM API Endpoints #### Features - **[Responses API](../../docs/response_api)** - Extract shared format mapping between Responses API and Chat Completions bridges - [PR #24417](https://github.com/BerriAI/litellm/pull/24417) - `use_chat_completions_api` flag for `openai/` models with custom `api_base` - [PR #25346](https://github.com/BerriAI/litellm/pull/25346) - `route_all_chat_openai_to_responses` global flag - [PR #25359](https://github.com/BerriAI/litellm/pull/25359) - Strip `custom_tool_call` namespace for all providers - [PR #26221](https://github.com/BerriAI/litellm/pull/26221) - **[Anthropic Messages API](../../docs/anthropic_unified)** - Map `reasoning_auto_summary` to `thinking.display` for native `/v1/messages` - [PR #25883](https://github.com/BerriAI/litellm/pull/25883) - Normalize reasoning effort with graceful degradation - [PR #26111](https://github.com/BerriAI/litellm/pull/26111) - **Memory API** - Add `/v1/memory` CRUD endpoints - [PR #26218](https://github.com/BerriAI/litellm/pull/26218) - Memory improvements v2 - [PR #26541](https://github.com/BerriAI/litellm/pull/26541) - **General** - Apply GPT-5 temperature validation in Responses API - [PR #24371](https://github.com/BerriAI/litellm/pull/24371) #### Bugs - **[Responses API](../../docs/response_api)** - Normalize bridged object field - [PR #26327](https://github.com/BerriAI/litellm/pull/26327) - **[Anthropic Messages API](../../docs/anthropic_unified)** - Preserve `anthropic_messages` call type for `/v1/messages` logging - [PR #26248](https://github.com/BerriAI/litellm/pull/26248) - **[Image API](../../docs/image_generation)** - Forward `litellm_params` to `validate_environment` for Vertex AI credentials in `image_edit` - [PR #26160](https://github.com/BerriAI/litellm/pull/26160) - Enforce multipart-only file inputs on image edit endpoints - [PR #26293](https://github.com/BerriAI/litellm/pull/26293) - Align image URL fetch with the validated HTTP client (Bedrock + token counter paths) - [PR #26272](https://github.com/BerriAI/litellm/pull/26272) - **[Vector Stores](../../docs/vector_stores)** - Restore BYOK key injection for vector store endpoints with team-scoped deployments - [PR #25746](https://github.com/BerriAI/litellm/pull/25746) - Respect object-level permissions for managed vector store endpoints - [PR #26351](https://github.com/BerriAI/litellm/pull/26351) - **Memory API** - JSONify metadata before Prisma writes on `/v1/memory` - [PR #26536](https://github.com/BerriAI/litellm/pull/26536) - **General** - Harden pass-through target URL construction - [PR #26467](https://github.com/BerriAI/litellm/pull/26467) --- ## Management Endpoints / UI #### Features - **Virtual Keys / Auth** - Refresh router after `POST /model/update` - [PR #26427](https://github.com/BerriAI/litellm/pull/26427) - Auto-add SSO team members to org on move (proxy admin only) - [PR #26377](https://github.com/BerriAI/litellm/pull/26377) - Apply team TPM/RPM + attribution for admins using `x-litellm-team-id` - [PR #26438](https://github.com/BerriAI/litellm/pull/26438) - Single-team DB fallback when JWT has no `team_id` - [PR #26418](https://github.com/BerriAI/litellm/pull/26418) - **UI** - "My User" tab on team info page - [PR #26520](https://github.com/BerriAI/litellm/pull/26520) - Send Invitation Email toggle on Users tab - [PR #25808](https://github.com/BerriAI/litellm/pull/25808) - UI setting to disable `/key/generate` for org admins - [PR #26442](https://github.com/BerriAI/litellm/pull/26442) - Sortable Model and TTFT columns on Spend Logs - [PR #26488](https://github.com/BerriAI/litellm/pull/26488) - Surface per-member budget cycle in Teams › Members tab - [PR #26207](https://github.com/BerriAI/litellm/pull/26207) - **Refactor** - Move projects management to enterprise package - [PR #25677](https://github.com/BerriAI/litellm/pull/25677) #### Bugs - **Virtual Keys / Auth** - Centralize `common_checks` to close authorization bypass - [PR #26279](https://github.com/BerriAI/litellm/pull/26279) - Tighten caller-permission checks on key route fields - [PR #26492](https://github.com/BerriAI/litellm/pull/26492) - Extend caller-permission checks to service-account + tighten raw-body acceptance - [PR #26493](https://github.com/BerriAI/litellm/pull/26493) - Enforce `upperbound_key_generate_params` on `/key/regenerate` - [PR #26340](https://github.com/BerriAI/litellm/pull/26340) - Preserve `service_account_id` in metadata on `/key/update` - [PR #26004](https://github.com/BerriAI/litellm/pull/26004) - Restrict `/global/spend/*` routes to admin roles - [PR #26490](https://github.com/BerriAI/litellm/pull/26490) - Harden team metadata handling in `/team/new` and `/team/update` - [PR #26464](https://github.com/BerriAI/litellm/pull/26464) - Extend request body parameter restrictions to cloud provider auth fields - [PR #26264](https://github.com/BerriAI/litellm/pull/26264) - Enforce format constraints on provider URL parameters - [PR #26287](https://github.com/BerriAI/litellm/pull/26287) - Bind RAG ingestion config to stored credential values - [PR #26512](https://github.com/BerriAI/litellm/pull/26512) - Broaden RAG ingestion credential cleanup to AWS endpoint/identity fields - [PR #26525](https://github.com/BerriAI/litellm/pull/26525) - Harden `/model/info` redaction for plural credential field names - [PR #26513](https://github.com/BerriAI/litellm/pull/26513) - **UI** - Stop injecting $0 cost on model edit - [PR #26001](https://github.com/BerriAI/litellm/pull/26001) --- ## AI Integrations ### Logging - **General** - Add `litellm_call_id` to `StandardLoggingPayload` and OTel span - [PR #26133](https://github.com/BerriAI/litellm/pull/26133) - **[Vertex AI Passthrough](../../docs/pass_through/vertex_ai)** - Log `:embedContent` and `:batchEmbedContents` responses - [PR #26146](https://github.com/BerriAI/litellm/pull/26146) ### Guardrails - **[Bedrock Guardrails](/docs/proxy/guardrails/bedrock)** - Use Bedrock OUTPUT source for `apply_guardrail` when scanning model responses - [PR #26144](https://github.com/BerriAI/litellm/pull/26144) - Dedupe post-call log entry when only `post_call` is configured - [PR #26474](https://github.com/BerriAI/litellm/pull/26474) - Hook mode + match redaction + streaming `request_data` for spend logs - [PR #25854](https://github.com/BerriAI/litellm/pull/25854), [PR #26266](https://github.com/BerriAI/litellm/pull/26266) - **LLM-as-a-Judge** - Ship LLM-as-a-Judge guardrail - [PR #26360](https://github.com/BerriAI/litellm/pull/26360) - **General** - Team-level guardrails and global policy guardrails can run together - [PR #26466](https://github.com/BerriAI/litellm/pull/26466) - Guardrail param handling in list and submission endpoints - [PR #26390](https://github.com/BerriAI/litellm/pull/26390) - Log `guardrail_information` on streaming post-call - [PR #26448](https://github.com/BerriAI/litellm/pull/26448) - Suppress deferred success log when post-call guardrail blocks - [PR #26528](https://github.com/BerriAI/litellm/pull/26528) --- ## Spend Tracking, Budgets and Rate Limiting - **Per-member budgets** - Individual team-member budgets - [PR #26208](https://github.com/BerriAI/litellm/pull/26208) - Track per-member total spend on team memberships - [PR #26195](https://github.com/BerriAI/litellm/pull/26195) - Fix per-team member budget bypass - [PR #26204](https://github.com/BerriAI/litellm/pull/26204) - **Rate limiting** - Reseed enforcement read path from DB on counter miss - [PR #26459](https://github.com/BerriAI/litellm/pull/26459) - **Budgets** - Align user and org budget spend checks with the atomic counter pattern - [PR #26182](https://github.com/BerriAI/litellm/pull/26182) - Reset budget windows failing due to Prisma `Json?` null filter - [PR #26346](https://github.com/BerriAI/litellm/pull/26346) --- ## MCP Gateway - **OAuth** - Harden OAuth `authorize`/`token` endpoints (BYOK + discoverable) - [PR #26274](https://github.com/BerriAI/litellm/pull/26274) - Share temporary MCP OAuth sessions across instances via Redis - [PR #26162](https://github.com/BerriAI/litellm/pull/26162), [PR #26318](https://github.com/BerriAI/litellm/pull/26318) - Align MCP OAuth proxy endpoints with per-server access policy - [PR #26516](https://github.com/BerriAI/litellm/pull/26516) - MCP broker OAuth endpoint access controls - [PR #26142](https://github.com/BerriAI/litellm/pull/26142) - **Permissions / routing** - Resolve team/key MCP permissions by name or alias - [PR #26338](https://github.com/BerriAI/litellm/pull/26338) - Split MCP routes into inference vs. management (unblocks Admin UI on `DISABLE_LLM_API_ENDPOINTS` nodes) - [PR #26367](https://github.com/BerriAI/litellm/pull/26367) - **Tool filtering** - Match tools with client-side namespace prefix in `mcp_semantic_tool_filter` - [PR #26117](https://github.com/BerriAI/litellm/pull/26117) --- ## Performance / Loadbalancing / Reliability improvements - **Routing** - Adaptive routing - [PR #26049](https://github.com/BerriAI/litellm/pull/26049) - Wildcard order fallback to higher-order deployments - [PR #25772](https://github.com/BerriAI/litellm/pull/25772) - **Prompt Compression** - First-class server-side prompt compression callback - [PR #25729](https://github.com/BerriAI/litellm/pull/25729) - **Reliability** - Fix `/health/readiness` 503 loop when DB is unreachable - [PR #26134](https://github.com/BerriAI/litellm/pull/26134) - **Developer ergonomics** - `--reload` flag for uvicorn hot reload (dev only) - [PR #25901](https://github.com/BerriAI/litellm/pull/25901) --- ## General Proxy Improvements - **Build / Docker** - Streamline `Dockerfile.non_root` build time - [PR #26055](https://github.com/BerriAI/litellm/pull/26055) - Use numeric UID 65534 in `docker.non_root` for K8s `runAsNonRoot` - [PR #26268](https://github.com/BerriAI/litellm/pull/26268) - Restore pre-uv Prisma cache path - [PR #26201](https://github.com/BerriAI/litellm/pull/26201) - **Migrations** - Opt-in v2 migration resolver - [PR #26194](https://github.com/BerriAI/litellm/pull/26194) - Freshness and destructive guards on migration workflow - [PR #26185](https://github.com/BerriAI/litellm/pull/26185) - **CI / Infra** - Migrate more CI jobs from CircleCI to GitHub Actions - [PR #26261](https://github.com/BerriAI/litellm/pull/26261) - CCI: cache, cleanup, anchors, install-path parity, Python 3.12, Ruby/Node pins - [PR #26286](https://github.com/BerriAI/litellm/pull/26286) - CircleCI config cleanup and consolidation - [PR #26226](https://github.com/BerriAI/litellm/pull/26226) - Speed up proxy unit tests and split `proxy-utils` into its own matrix entry - [PR #26150](https://github.com/BerriAI/litellm/pull/26150) - Remove CCI/GHA test duplication and semantically shard proxy DB tests - [PR #26356](https://github.com/BerriAI/litellm/pull/26356) - Standalone `create-release-branch` workflow + `contents:write` permission - [PR #26342](https://github.com/BerriAI/litellm/pull/26342), [PR #26359](https://github.com/BerriAI/litellm/pull/26359) - Supply-chain guard to block fork PRs that modify dependencies - [PR #26511](https://github.com/BerriAI/litellm/pull/26511) - Use Postgres sidecar instead of shared DB for `auth_ui_unit_tests` - [PR #26141](https://github.com/BerriAI/litellm/pull/26141) - Fix `e2e_ui_testing` stale-bundle issue on Ubuntu (`cp -r` merge semantics) - [PR #26047](https://github.com/BerriAI/litellm/pull/26047) - Apply black formatting to fix CI lint failures - [PR #26140](https://github.com/BerriAI/litellm/pull/26140) - **Test stability** - Stabilize spend-accuracy tests + patch Redis buffer data-loss path - [PR #26270](https://github.com/BerriAI/litellm/pull/26270) - Stabilize spend-accuracy test transport flakes - [PR #26290](https://github.com/BerriAI/litellm/pull/26290) - Deflake spend-tracking tests - [PR #26349](https://github.com/BerriAI/litellm/pull/26349) - Drain logging worker in `test_router_caching_ttl` to fix flakiness - [PR #26355](https://github.com/BerriAI/litellm/pull/26355) - Isolate `master_key`/`prisma_client` module globals between proxy tests - [PR #26362](https://github.com/BerriAI/litellm/pull/26362) - **Packaging / dependencies** - Bump vulnerable dependencies - [PR #26365](https://github.com/BerriAI/litellm/pull/26365) - Declare MIT license in `litellm-proxy-extras` metadata - [PR #26369](https://github.com/BerriAI/litellm/pull/26369) - Declare proprietary license in `litellm-enterprise` metadata - [PR #26457](https://github.com/BerriAI/litellm/pull/26457) - **UI** - Fetch button ignores active filters on Request Logs page - [PR #25788](https://github.com/BerriAI/litellm/pull/25788) - Stale filters applied after sort/page/time change on Request Logs - [PR #25789](https://github.com/BerriAI/litellm/pull/25789) - **Misc** - Replace substring check with `startswith` in `is_model_gpt_5_model` - [PR #25793](https://github.com/BerriAI/litellm/pull/25793) --- ## Documentation Updates - Add missing observability integrations to View All page - [PR #24420](https://github.com/BerriAI/litellm/pull/24420) - Clarify `x-litellm-model-group` vs. provider model id in proxy docs - [PR #25497](https://github.com/BerriAI/litellm/pull/25497) - Gemini 3 thinking_level defaults and release note - [PR #25842](https://github.com/BerriAI/litellm/pull/25842) - Align fenced code block padding on blog and doc pages - [PR #25932](https://github.com/BerriAI/litellm/pull/25932) - Add supported providers to prompt caching doc - [PR #26124](https://github.com/BerriAI/litellm/pull/26124) - Remove `docs/my-website`, point contributors to `BerriAI/litellm-docs` - [PR #26454](https://github.com/BerriAI/litellm/pull/26454) --- ## New Contributors - @dongyu-turo made their first contribution in [#24164](https://github.com/BerriAI/litellm/pull/24164) - @Alpha-Zark made their first contribution in [#25672](https://github.com/BerriAI/litellm/pull/25672) - @vinhphamhuu-ct made their first contribution in [#25767](https://github.com/BerriAI/litellm/pull/25767) - @Bytechoreographer made their first contribution in [#25788](https://github.com/BerriAI/litellm/pull/25788) - @BraulioV made their first contribution in [#25793](https://github.com/BerriAI/litellm/pull/25793) - @Vigilans made their first contribution in [#25883](https://github.com/BerriAI/litellm/pull/25883) - @nhyy244 made their first contribution in [#26110](https://github.com/BerriAI/litellm/pull/26110) - @sakenuGOD made their first contribution in [#26117](https://github.com/BerriAI/litellm/pull/26117) - @Michael-RZ-Berri made their first contribution in [#26124](https://github.com/BerriAI/litellm/pull/26124) - @anmolg1997 made their first contribution in [#26228](https://github.com/BerriAI/litellm/pull/26228) **Full Changelog**: https://github.com/BerriAI/litellm/compare/v1.83.10-stable...v1.83.14-stable --- ## 04/27/2026 * New Models / Updated Models: 29 * LLM API Endpoints: 18 * Management Endpoints / UI: 23 * AI Integrations (Logging / Guardrails): 11 * Spend Tracking, Budgets and Rate Limiting: 6 * MCP Gateway: 8 * Performance / Loadbalancing / Reliability improvements: 5 * General Proxy Improvements: 27 * Documentation Updates: 6 ## Related pages - [v1.84.0](https://docs.litellm.ai/release_notes/v1.84.0/v1-84-0.md) - [v1.83.10](https://docs.litellm.ai/release_notes/v1.83.10/v1-83-10.md)