aiohttp

This release introduces document reranking and standard model exception types. It also includes fixes for filtering invalid tool calls and avoiding repeated service tier usage.

This release introduces new features like `reasoning_effort` parameter and content-block-centric streaming, alongside several bug fixes and dependency updates.

This release primarily focuses on dependency updates across various components, including langsmith, vcrpy, and pytest, alongside introducing package version tracking for tracing metadata.

This release primarily updates the dependency version for aiohttp within the langchain_v1 library.

This release introduces Docker image signature verification using cosign for enhanced security and includes numerous bug fixes and feature enhancements across various components like Prometheus, aiohttp, Vertex AI, Anthropic adapter, and the UI.

This release introduces Docker image signature verification using cosign and includes numerous bug fixes and feature enhancements across the proxy, UI, and MCP components. Key updates involve improved security, better management endpoints, and enhanced routing capabilities.

This release introduces Docker image signature verification using cosign and includes numerous bug fixes and feature enhancements across various components like Prometheus, aiohttp, Helm, Vertex AI, Anthropic adapter, and the UI. Notably, the ai-gateway handler has been removed in favor of litellm-core.

This release focuses primarily on bug fixes across various integrations, including recovery from zombie processes, improved pricing application, and fixes related to rate limiting and authentication flows. Several recent fixes were also reverted.

This release focuses heavily on stability, security fixes (including CVEs), and extensive UI/UX improvements across logging, teams, and authentication flows. New features include support for Websearch interception hooks and expanded model support for Anthropic, Sarvam AI, and Perplexity.

This release focuses heavily on stability, bug fixes across various providers (including Bedrock, Cohere, Vertex AI), and significant improvements to the LiteLLM UI and OAuth2 support. New features include Websearch interception and support for Sarvam AI and Perplexity Research API.

This release introduces significant feature additions, including new provider support (OVHCloud, CompactifAI, Sambarova models), enhanced guardrail capabilities (tool-permission guardrail, Bedrock Guardrails), and major performance optimizations. Numerous bug fixes address issues across Bedrock, Cohere, Vertex AI, and proxy configurations.

This release focuses heavily on enhancing the MCP Gateway with new features like custom cost tracking, tool testing, and improved security/UI fixes across the platform. Key updates include better Bedrock guardrail handling and various bug fixes related to logging and configuration parsing.

This release introduces significant feature enhancements, including JWT authentication sync, expanded guardrails support (Azure, Bedrock), and new provider integrations like deepseek-ai and XInference. Numerous bug fixes address stability, logging, and dependency issues across various components, especially around MCP.

This release introduces significant feature enhancements across Azure, VertexAI, and Ollama integrations, alongside major improvements to security scanning, UI management (including health checks and organization settings), and expanded support for new models and API endpoints.

This release introduces significant feature enhancements across model support (including new pricing and reasoning capabilities), UI improvements for management (MCP), and numerous bug fixes related to streaming, provider integrations (VertexAI, Deepgram), and internal routing logic. Key additions include Lasso Guardrail support and expanded Responses API functionality.

This release introduces significant enhancements around Multi-Cloud Proxy (MCP) management via the UI and API, adds support for the Deepgram provider, and includes various performance and stability fixes across routing and logging.

This release introduces significant feature enhancements across multiple providers, including Gemini parallel tool calls, Azure image edits, and Bedrock video support. Numerous bug fixes address issues related to dependencies, provider configurations, and UI stability.

This release introduces several new features, including expanded tool support for VertexAI and Bedrock, and enhanced compatibility with Gemini and Azure services. It also includes important bug fixes related to dependency versions and data tracking.

This release introduces experimental support for HTTPX2 clients and enhances CI workflow templates. It also addresses a dependency issue by requiring a patched aiohttp version for Python 3.10 and above.

This release introduces support for aiohttp and includes several API updates, alongside bug fixes for binary response parsing and test suite improvements.

This release introduces extensive new tracing support across LLM frameworks (Bedrock, CrewAI, LiteLLM, OpenAI Agents) and LangGraph, alongside DBM propagation for PyMongo. Several deprecations for older tracing standards and bug fixes across IAST, CI Visibility, and core integrations are also included.

This release introduces extensive new tracing features for LLM frameworks like Bedrock, LangGraph, CrewAI, and LiteLLM, alongside DBM propagation support for PyMongo. Several deprecated features are marked for removal in v4.0.0, and numerous bug fixes address stability and correctness across CI Visibility, IAST, and various integrations.

This release focuses heavily on cleaning up deprecated tracing configurations, removing legacy metrics/sampling methods from LangChain and OpenAI integrations, and enforcing a single global tracer instance. A new feature includes automatic instrumentation for the Valkey package.

This release brings various bug fixes across attention mechanisms, hardware acceleration (ROCm, DirectML, Ascend NPU), and dependency updates. Key feature additions include support for ancestral samplers and Lumina2 system prompt integration.