Skip to main content

Decisions API is here. Call Jev, Clef or any decision model through one /v1/decisions endpoint. →

Docs

HomeGet started with LiteLLM

Build

AI GatewayRoute, control, and observe LLM traffic MCP GatewayGive agents governed access to tools Agent GatewayRegister and invoke A2A agents Auto RouterSend each request to the best model Python SDKCall 100+ LLMs with one interface

Monitor

LensTrace agent swarms and find what to improve AI Gateway - Logging & ObservabilityLogs, spend, and callbacks for every request

Self-Hosted Coding Agents

MoyaiOpen source coding agent for background work

Docs Integrations Enterprise Learn Changelog Blog Rust

All Posts

2026

Introducing Microsoft 365 Copilot in LiteLLM
Introducing litellm-core: the first step toward a leaner SDK
Claude Haiku 5.5 API Pricing and Day 0 Support on LiteLLM
Open Sourcing Moyai: Self-Hosted Cloud Coding Agent
Day 0 Support: Nano Banana 2.1
How LiteLLM Lens finds repeated failures across 1,000s of agent traces
How we cut time to first byte by 94% for long prompts
How we built our own internal Devin in 2 days
Adding Self-hosted Auto Router Classifiers: Laya & Nimble
Launching LiteLLM Lens
How we cut LiteLLM's Redis round trips per request by 64%
How we made the LiteLLM Usage page 120x faster
Day 0 Support: GPT-6.1 Sol
Learn the LiteLLM gateway with a guided course
Day 0 Support: Claude Sonnet 5.5
Introducing the LiteLLM ROI Calculator
Introducing LiteAgents
September Townhall Updates: 583 Bug Fixes, OCR on Rust, and 83.5% Coverage
Day 0 support: Gemini 3.8 Flash TTS and Flash-Lite TTS
Introducing LiteAdmin MCP
Day 0 Support: GPT-6 Sol and GPT-6 Luna
Day 0 Support: Claude Opus 5.5
Getting Started with Fireworks AI on LiteLLM
Day 0 Support: Xiaomi MiMo V2.6
Claude Code server-side auto mode through LiteLLM
Day 0 Support: Grok 4.7
JEV Classifier: 5.43x as Fast as Haiku, 96% Lower Cost
TypeSafe Jev on LiteLLM
Auto-Router: Switching Tiers Without Encrypted Content Failures
Day 0 Support: Qwen3.8-Omni-Flash
Reduce agent context with TypeSafe Jev and LiteLLM
Auto Router: 5 Improvements for Cost, Speed, and Team Control
September Townhall: Product + Roadmap Updates
Auto Router: Maximize Quality & Savings with Our Fuse LLM Classifier
OCR uses Rust by default starting with v1.102.0-rc.1
Incident Report: Retry Breadcrumb Memory Growth Causing OOM on v1.100.0
Auto Router: 45% Lower Cost on 25 SWE-bench Tasks
How Pfizer Improved LiteLLM Gateway Performance and Resiliency at Scale
Auto-Router Updates: Harness-Aware Routing
Secure shared AI agents with identity-aware access and spend controls
AutoRouter: Tune Heuristics for Your Traffic
Our Updated SOC 2 Type 2 Report Is Available
Subtask-Specific Routing: Same Quality, 46% Less Cost
AutoRouter Per-Hop Compression: Cut LLM Classifier Costs Another 32%
Auto-Router: Escalate a Task That Gets Stuck
Day 0 Support: GPT-6 Astra
Day 0 support: Meta Muse Spark 1.3
Day 0 support: Gemini 3.8 Flash
Introducing AutoRouter Heuristic v2: 27% More Tasks Solved at 45% Lower Cost
Introducing LiteLLM Fusion: 56% More Tasks Solved Than Fable 5
Auto-Router: Route on Context Size and Modality
Day 0 Support: Claude Fable 5.1
August Townhall Updates: Security, Stability, and Product
Shadow Evaluations: Test the Auto-Router on Your Own Production Traffic
Day 0 support: Gemini 3.7 Flash
August Townhall: Product + Roadmap Updates
51% Cost Savings Reported From a Live Production Deployment
Auto Router: Opus level quality at up to 27% lower cost
AutoRouter: Easy Visibility to Your Savings
AutoRouter: 1 Click Deploy
Auto Router v1.97: usage benchmarks and better quality for lower cost
Prompt Caching Works with Auto Router
Cut 75% Claude Code cost with near frontier model quality
July Townhall Updates: 38 Security Fixes, 317 Bug Fixes, and Autorouter V2
Day 0 Support: Claude Opus 5
Benchmarking the LiteLLM Rust AI Gateway: Overhead, Memory, and Cost
Announcing Router Plugins: Customize Routing Signals
Auto Router v2: one router for complexity, semantic, and adaptive routing
Incident Report: Prompt Cache Invalidation for Claude Code on Bedrock Invoke
July stability update: hardening MCP auth and cutting pass-through memory
July Townhall: Product + Roadmap Updates
Day 0 Support: GPT-5.6 (Sol, Terra, Luna)
5 ways to cut Claude Code costs with LiteLLM
Day 0 Support: Claude Sonnet 5
LiteLLM × Headroom: Use 60-95% fewer tokens with Claude Code
June Townhall Updates: 94 Bug Fixes, OCR + Realtime are in Rust, and a Zero-Regression Commitment
Swap OpenAI Code Interpreter for E2B/OpenSandbox
Migrating LiteLLM to Rust - Building the Fastest and Litest AI Gateway
LiteLLM version support: focusing on the four most recent stable lines
Semantic Caching on Valkey and AWS ElastiCache
June Townhall: Product + Roadmap Updates
June Stability Update: We're Making Stability a First-Class Citizen at LiteLLM
Day 0 Support: Claude Fable 5
A Unified Agent Control Plane
Announcing LiteLLM x Microsoft ASSERT
LiteLLM Labs: Announcing Lite-Harness SDK — Unified API for Claude Code, Codex, and Pi AI
Fixed in 1.84.0+ - Version Update: Authentication Bypass via Host Header Injection (GHSA-4xpc-pv4p-pm3w)
Day 0 Support: Claude Opus 4.8
How we built a background agent to cover 30% of our backlog
May Townhall Updates: Security Hardening, Release Versioning, and the Agent Platform
DAY 0 Support: Gemini 3.5 Flash on LiteLLM
Google AI Studio Managed Agents on LiteLLM
May Townhall: Product + Roadmap Updates
Announcing Componentized Deployments
Security Update: Mistral AI PyPI Supply Chain Attack — LiteLLM Not Impacted
LiteLLM Managed Agents Platform — Alpha Now Open for Public Preview
Security Update: CVE-2026-42208 in LiteLLM Proxy
Incident Report: Prisma DB Reconnect Blocks the Event Loop and Kills Liveliness
LiteLLM release versioning is changing: standard names, MINOR for weekly, PATCH for hotfixes
Gemini Embedding 2 (GA): Multimodal Embeddings on LiteLLM
Day 0 Support: GPT-5.5 and GPT-5.5 Pro
Security Update: CVE-2026-30623 — Command Injection via Anthropic's MCP SDK
LiteLLM × Akto: Model-Based Detection Alongside Built-in Guardrails
Day 0 Support: Claude Opus 4.7
Making the AI Gateway Resilient to Redis Failures
April Townhall Updates: CI/CD v2, Stability, and Product Roadmap
Security Update: Vulnerability Disclosures and Ongoing Hardening
April Townhall: Security + Product Roadmap
Announcing CI/CD v2 for LiteLLM
LiteLLM + Vanta: SOC 2 Type 2 and ISO 27001 Recertification
Security Townhall Updates
Security Update: Suspected Supply Chain Incident
Incident Report: Guardrail logging exposed secret headers in spend logs and traces
Day 0 Support: GPT-5.4-mini and GPT-5.4-nano
New Video Characters, Edit and Extension API support
Realtime WebRTC HTTP Endpoints
Day 0 Support: GPT-5.4
DAY 0 Support: Gemini 3.1 Flash Lite Preview on LiteLLM
Incident Report: Cache Eviction Closes In-Use httpx Clients
Day 0 Support: GPT-5.3-Codex
Incident Report: Encrypted Content Failures in Multi-Region Responses API Load Balancing
Incident Report: Wildcard Blocking New Models After Cost Map Reload
Incident Report: SERVER_ROOT_PATH regression broke UI routing
DAY 0 Support: Gemini 3.1 Pro on LiteLLM
Incident Report: vLLM Embeddings Broken by encoding_format Parameter
Day 0 Support: Claude Sonnet 4.6
Incident Report: Invalid beta headers with Claude Code
Day 0 Support: MiniMax-M2.5
Incident Report: Invalid model cost map on main
Your Middleware Could Be a Bottleneck
Improve release stability with 24 hour load tests
Day 0 Support: Claude Opus 4.6
Achieving Sub-Millisecond Proxy Overhead

2025

DAY 0 Support: Gemini 3 Flash on LiteLLM
Day 0 Support: Claude 4.5 Opus (+Advanced Features)
DAY 0 Support: Gemini 3 on LiteLLM
Gemini Embedding 2 Preview: Multimodal Embeddings on LiteLLM

Tags

A

adaptive1
admin-ui1
advanced features1
agent-tracing2
agents12
ai-gateway15
akto1
announcement9
anthropic14
architecture1
auth1
auto-router4

B

bedrock1
benchmarks12
budgets1

C

caching4
case-study1
characters1
ci-cd1
claude12
claude haiku1
claude-code4
code interpreter1
codex1
completion7
complexity-router19
compliance2
compression1
context1
cost13
cost-optimization2
customer-story1

D

dashscope1
database1
day 0 support27
docker1

E

e2b1
effort1
elasticache1
embeddings3
engineering29
evals1

F

fable 51
fable 5.11
fastapi1
finops1
fireworks1

G

gemini12
gemini enterprise agent platform1
gemini-3.7-flash1
gemini-3.8-flash1
gemini-3.8-flash-tts1
getting-started1
google-ai-studio1
gpt-5.3-codex1
gpt-5.41
gpt-5.4-mini1
gpt-5.4-nano1
gpt-5.51
gpt-5.5-pro1
gpt-5.61
gpt-5.6-luna1
gpt-5.6-sol1
gpt-5.6-terra1
gpt-63
gpt-6-astra1
gpt-6-luna1
gpt-6-sol1
gpt-6.1-sol1
grok1
grok-4.71
guardrails4

H

haiku 5.51
harnesses1
headroom2

I

ideas1
image generation1
incident-report14
infrastructure3
interactions1

J

jev1

K

kubernetes1

L

lap1
lens2
lite-harness1
litellm1
litellm-labs1
llm1
llms8
load-balancing1

M

M2.51
managed-agents1
mcp4
memory1
meta1
middleware1
mimo1
mimo-v2.61
minimax1
mistral1
model-access1
multimodal2
muse-spark-1.31

N

nano banana1

O

observability2
ocr1
open-source1
openai9
opensandbox1
opus 4.61
opus 4.71
opus 4.81
opus 51
opus 5.51

P

packaging1
partnership2
performance11
plugins1
postgres1
pricing1
prisma1
product20
programmatic tool calling1
prompt-caching1
providers1
proxy8

Q

quality1
qwen1
qwen3.8-omni-flash1

R

realtime1
redis4
release2
releases1
reliability11
responses-api2
router1
routing24
rust3
rust-migration3

S

sandbox1
scaling1
sdk2
security19
semantic cache1
semantic-router1
slack1
sonnet 4.61
sonnet 51
sonnet 5.51
stability5
supernova1
supply-chain1
support1

T

testing2
text to speech1
tool search1
townhall12
tutorial1
typesafe1

U

ui3

V

valkey1
vertex ai2
videos1
vllm1

W

webrtc1

X

xai1
xiaomi1

One API for every model, in your code or behind your own gateway.

Get the latest LiteLLM updates

New model support, product launches, and release notes, in your inbox.

Product

Gateway quickstart
Python SDK
Production deployment
MCP Gateway
Agent Gateway
Rust AI Gateway (beta)
Enterprise

Resources

Blog
Changelog
Agent resources
llms.txt
Trust Center

Community

GitHub
Discord
Slack
YouTube
X
LinkedIn

Company

litellm.ai
Talk to sales
Careers

© 2026 LiteLLM

SOC 2 Type II MIT license