Berri AI

The fastest way to take your LLM app to production

GitHub 73 repos 1.2k followers

Explained projects

Yet to be explained

moyai
A self-hosted coding agent for background work - works with Claude Code, Codex, OpenCode & 100+ providers
Python55 stars
Explain
liteagents
A provider-independent agent SDK with the same query() interface as the Claude Agent SDK, allowing you to use the right model for every turn.
Python36 stars
Explain
litespeed
Litespeed — a local coding agent with multi-model workflows for your terminal and browser
TypeScript28 stars
Explain
ai-gateway-bench
Benchmarks for evaluating ai-gateways
Python18 stars
Explain
litellm-admin-agent
Internal LiteLLM admin assistant for BerriAI Slack and gateway Agents
Python1 stars
Explain
shipr
Shiprr: minimal agentic coding CLI with smart LiteLLM routing
Rust1 stars
Explain
litellm-rust-challenge
LiteLLM Rust Challenge
1 stars
Explain
litellm-lens-codex-integration
Codex session tracing for LiteLLM Lens. A BerriAI-maintained public preview.
Python0 stars
Explain
deeplite
Python0 stars
Explain
agentchat
A unified Python SDK for connecting ai agents to Slack, Discord, Teams, Email
Python0 stars
Explain
litellm-roi-calculator
Local dashboard comparing LiteLLM gateway spend with estimated engineering hours from GitHub pull requests.
Python0 stars
Explain
liteadmin-mcp
Standalone MCP connector for administering a LiteLLM gateway: keys, model deployments, teams, budgets and spend.
Python0 stars
Explain
litellm-bench
TypeScript0 stars
Explain
litellm-mongodb
Python0 stars
Explain
codex-litellm-stack
Codex + LiteLLM stack: set up with an agent or a script
Shell0 stars
Explain
litellm-readme-7f3c91e4
Private-link style LiteLLM team directory
HTML0 stars
Explain