LiteLLM is an open-source LLM Gateway with 18K+ stars on GitHub and trusted by companies like Rocket Money, Samsara, Lemonade, and Adobe. LiteLLM provides an open source Python SDK and Python FastAPI Server that allows calling 100+ LLM APIs (Bedrock, Azure, OpenAI, VertexAI, Cohere, Anthropic) in the OpenAI format
We have raised a $1.6M seed round from Y Combinator, Gravity Fund and Pioneer Fund
AI Gateway to provide model access, fallbacks and spend tracking across 100+ LLMs.
What customers say
“LiteLLM has let my team provide the latest LLM models to our users, usually within a day of them being released… it has saved us months of work.”
CustomerLiteLLM customer
“Single access source with consistencyLiteLLM gives NVIDIA engineers a single, consistent way to access more than 100 AI model endpoints across cloud providers, open source deployments, and internal NVIDIA services.”
CustomerLiteLLM customer
“LiteLLM gives NVIDIA engineers a single, consistent way to access more than 100 AI model endpoints across cloud providers, open source deployments, and internal NVIDIA services.”
CustomerLiteLLM customer
Security and compliance
As reported by LiteLLM. Ask for audit reports before you buy.
SOC 2 Type ICertified
SOC 2 Type IIIn progress
ISO/IEC 27001Certified
ISO 22301Compliant
TLS/SSL (TLS 1.2 and TLS 1.3)Compliant
NaCl SecretBox (XSalsa20-Poly1305) encryption for API keys, credentials, and config secretsCompliant
CodeQLCompliant
grype vulnerability scanning on Docker imagesCompliant
Self-hosted deployment with no data or telemetry stored on LiteLLM servers
No personal data collected or transmitted for self-hosted users
Minimal session cookies (JWT) for Admin UI authentication; no third-party cookies
Audit logs for auth, creation, and update events
Vulnerability disclosure policy
Cybersecurity incident management program with incident response policy
Partial data-at-rest encryption: API keys, credentials, and config secrets encrypted via NaCl SecretBox; spend/audit logs and cache data not encrypted
Is this your company?
Claim the profile to keep facts current, answer buyers and submit evidence for the next ranking review.
LiteLLM v1.93.0 Release: New Model Support, Meta Provider, and MCP Enhancements
LiteLLM released version 1.93.0, introducing support for new models including OpenAI GPT-5.6, xAI Grok-4.5, OpenAI Realtime 2.1, and Google Cloud Chirp 3. The update includes a new OpenAI-compatible 'Meta' model provider for muse-spark-1.1 and adds client-forwarded MCP credential support with true_passthrough and oauth_delegate modes. The dashboard was migrated to Base UI primitives alongside a smarter complexity router. Notable changes include throttling API keys instead of revoking access upon hitting spend limits and moving OpenTelemetry error detail keys to the litellm namespace.
