- February 2 - Achieving Sub-Millisecond Proxy Overhead
- February 5 - Day 0 Support: Claude Opus 4.6
- February 6 - Improve release stability with 24 hour load tests
- February 7 - Your Middleware Could Be a Bottleneck
- February 10 - Incident Report: Invalid model cost map on main
- February 12 - Day 0 Support: MiniMax-M2.5
- February 16 - Incident Report: Invalid beta headers with Claude Code
- February 17 - Day 0 Support: Claude Sonnet 4.6
- February 18 - Incident Report: vLLM Embeddings Broken by encoding_format Parameter
- February 19 - DAY 0 Support: Gemini 3.1 Pro on LiteLLM
- February 21 - Incident Report: SERVER_ROOT_PATH regression broke UI routing
- February 23 - Incident Report: Wildcard Blocking New Models After Cost Map Reload
- February 24 - Incident Report: Encrypted Content Failures in Multi-Region Responses API Load Balancing
- February 24 - Day 0 Support: GPT-5.3-Codex
- February 27 - Incident Report: Cache Eviction Closes In-Use httpx Clients
- March 3 - DAY 0 Support: Gemini 3.1 Flash Lite Preview on LiteLLM
- March 5 - Day 0 Support: GPT-5.4
- March 12 - Realtime WebRTC HTTP Endpoints
- March 16 - New Video Characters, Edit and Extension API support
- March 17 - Day 0 Support: GPT-5.4-mini and GPT-5.4-nano
- March 18 - Incident Report: Guardrail logging exposed secret headers in spend logs and traces
- March 24 - Security Update: Suspected Supply Chain Incident
- March 27 - Security Townhall Updates
- March 30 - LiteLLM + Vanta: SOC 2 Type 2 and ISO 27001 Recertification
- March 30 - Announcing CI/CD v2 for LiteLLM
- April 2 - April Townhall: Security + Product Roadmap
- April 3 - Security Update: Vulnerability Disclosures and Ongoing Hardening
- April 10 - April Townhall Updates: CI/CD v2, Stability, and Product Roadmap
- April 11 - Making the AI Gateway Resilient to Redis Failures
- April 16 - Day 0 Support: Claude Opus 4.7
- April 21 - LiteLLM × Akto: Model-Based Detection Alongside Built-in Guardrails
- April 21 - Security Update: CVE-2026-30623 — Command Injection via Anthropic's MCP SDK
- April 24 - Day 0 Support: GPT-5.5 and GPT-5.5 Pro
- April 24 - Gemini Embedding 2 (GA): Multimodal Embeddings on LiteLLM
- April 28 - LiteLLM release versioning is changing: standard names, MINOR for weekly, PATCH for hotfixes
- April 29 - Incident Report: Prisma DB Reconnect Blocks the Event Loop and Kills Liveliness
- April 29 - Security Update: CVE-2026-42208 in LiteLLM Proxy
- May 8 - LiteLLM Managed Agents Platform — Alpha Now Open for Public Preview
- May 12 - Security Update: Mistral AI PyPI Supply Chain Attack — LiteLLM Not Impacted
- May 18 - Announcing Componentized Deployments
- May 19 - May Townhall: Product + Roadmap Updates
- May 19 - Google AI Studio Managed Agents on LiteLLM
- May 19 - DAY 0 Support: Gemini 3.5 Flash on LiteLLM
- May 26 - May Townhall Updates: Security Hardening, Release Versioning, and the Agent Platform
- May 27 - How we built a background agent to cover 30% of our backlog
- May 28 - Day 0 Support: Claude Opus 4.8
- June 1 - Fixed in 1.84.0+ - Version Update: Authentication Bypass via Host Header Injection (GHSA-4xpc-pv4p-pm3w)
- June 2 - LiteLLM Labs: Announcing Lite-Harness SDK — Unified API for Claude Code, Codex, and Pi AI
- June 3 - Announcing LiteLLM x Microsoft ASSERT
- June 10 - A Unified Agent Control Plane
- June 10 - Day 0 Support: Claude Fable 5
- June 15 - June Stability Update: We're Making Stability a First-Class Citizen at LiteLLM
- June 16 - June Townhall: Product + Roadmap Updates
- June 17 - Semantic Caching on Valkey and AWS ElastiCache
- June 20 - LiteLLM version support: focusing on the four most recent stable lines
- June 22 - Migrating LiteLLM to Rust - Building the Fastest and Litest AI Gateway
- June 23 - Swap OpenAI Code Interpreter for E2B/OpenSandbox
- June 26 - June Townhall Updates: 94 Bug Fixes, OCR + Realtime are in Rust, and a Zero-Regression Commitment
- June 30 - LiteLLM × Headroom: Use 60-95% fewer tokens with Claude Code
- June 30 - Day 0 Support: Claude Sonnet 5
- July 4 - 5 ways to cut Claude Code costs with LiteLLM
- July 9 - Day 0 Support: GPT-5.6 (Sol, Terra, Luna)
- July 9 - July Townhall: Product + Roadmap Updates
- July 11 - July stability update: hardening MCP auth and cutting pass-through memory
- July 13 - Incident Report: Prompt Cache Invalidation for Claude Code on Bedrock Invoke
- July 13 - Auto Router v2: one router for complexity, semantic, and adaptive routing
- July 17 - Announcing Router Plugins: Customize Routing Signals
- July 22 - Benchmarking the LiteLLM Rust AI Gateway: Overhead, Memory, and Cost
- July 24 - Day 0 Support: Claude Opus 5
- July 24 - July Townhall Updates: 38 Security Fixes, 317 Bug Fixes, and Autorouter V2
- July 27 - Cut 75% Claude Code cost with near frontier model quality
- July 31 - Prompt Caching Works with Auto Router
- August 4 - Auto Router v1.97: usage benchmarks and better quality for lower cost
- August 5 - AutoRouter: 1 Click Deploy
- August 6 - AutoRouter: Easy Visibility to Your Savings
- August 8 - Auto Router: Opus level quality at up to 27% lower cost
- August 10 - 51% Cost Savings Reported From a Live Production Deployment
- August 10 - August Townhall: Product + Roadmap Updates
- August 13 - Day 0 support: Gemini 3.7 Flash
- August 18 - Shadow Evaluations: Test the Auto-Router on Your Own Production Traffic
- August 27 - August Townhall Updates: Security, Stability, and Product
- September 1 - Day 0 Support: Claude Fable 5.1
- September 1 - Auto-Router: Route on Context Size and Modality
- September 1 - Introducing LiteLLM Fusion: 56% More Tasks Solved Than Fable 5
- September 2 - Introducing AutoRouter Heuristic v2: 27% More Tasks Solved at 45% Lower Cost
- September 2 - Day 0 support: Gemini 3.8 Flash
- September 2 - Day 0 support: Meta Muse Spark 1.3
- September 3 - Day 0 Support: GPT-6 Astra
- September 4 - Auto-Router: Escalate a Task That Gets Stuck
- September 5 - AutoRouter Per-Hop Compression: Cut LLM Classifier Costs Another 32%
- September 7 - Subtask-Specific Routing: Same Quality, 46% Less Cost
- September 8 - Our Updated SOC 2 Type 2 Report Is Available
- September 8 - AutoRouter: Tune Heuristics for Your Traffic
- September 8 - Secure shared AI agents with identity-aware access and spend controls
- September 10 - Auto-Router Updates: Harness-Aware Routing
- September 11 - How Pfizer Improved LiteLLM Gateway Performance and Resiliency at Scale
- September 11 - Auto Router: 45% Lower Cost on 25 SWE-bench Tasks
- September 12 - Incident Report: Retry Breadcrumb Memory Growth Causing OOM on v1.100.0
- September 13 - OCR uses Rust by default starting with v1.102.0-rc.1
- September 14 - Auto Router: Maximize Quality & Savings with Our Fuse LLM Classifier
- September 16 - September Townhall: Product + Roadmap Updates
- September 17 - Auto Router: 5 Improvements for Cost, Speed, and Team Control
- September 18 - Reduce agent context with TypeSafe Jev and LiteLLM
- September 18 - Day 0 Support: Qwen3.8-Omni-Flash
- September 18 - Auto-Router: Switching Tiers Without Encrypted Content Failures
- September 20 - TypeSafe Jev on LiteLLM
- September 20 - JEV Classifier: 5.43x as Fast as Haiku, 96% Lower Cost
- September 21 - Day 0 Support: Grok 4.7
- September 21 - Claude Code server-side auto mode through LiteLLM
- September 22 - Day 0 Support: Xiaomi MiMo V2.6
- September 22 - Day 0 Support: Claude Opus 5.5
- September 22 - Day 0 Support: GPT-6 Sol and GPT-6 Luna
- September 23 - Day 0 support: Gemini 3.8 Flash TTS and Flash-Lite TTS