ArchestackBest of Open-Source AI: List on GitHub ↗3 stars (opens in a new tab)

#3434th of 48 in Agent frameworks and SDKs

aisuite

Python library with one chat API across LLM providers, plus an agents layer

Documentation ↗ (opens in a new tab)Repository on GitHub ↗ (opens in a new tab)
Stars
16.3k
License
MIT
Last commit
Sep 2026
Last release
Jul 2026
Language
Python

Missed the latest GitHub check; these facts may be older.

Overview

aisuite is a Python library that exposes an OpenAI-style Chat Completions interface across providers such as OpenAI, Anthropic, Google, Mistral, Ollama and AWS, selected with a "provider:model" string. On top of that it adds tool calling from plain Python functions with max_turns, an Agents API with a Runner, toolkits for files, git and shell, and MCP server support. The desktop app OpenWorker now lives in a separate repository.

Who it is for: Python developers building LLM apps or agents that must work across providers

Strengths

  • Switch providers by changing one model string, e.g. openai:gpt-4o to anthropic:...
  • Plain Python functions become tools; schemas generated and calls executed automatically via max_turns
  • Agents API includes tool policies, state stores (memory, file, Postgres) and tracing
  • Native MCP client support and streaming, with async variant

Weaknesses

  • Library only, no server, proxy, UI, rate limiting or caching
  • Streaming is limited to some providers and cannot be combined with max_turns
  • Provider SDKs and API keys are installed and managed separately by you
  • OpenWorker app moved out; only an archived snapshot remains in this repo

What it needs

  • no GPU
  • Needs Provider API keys, Provider SDKs (optional extras), Node/npx for some MCP servers
  • Models: OpenAI, Anthropic, Google, Mistral, Hugging Face

Also in Agent frameworks and SDKs

See all 48
Also in Agent frameworks and SDKs
RankProjectScore
1ppt-masterAgent skill that turns documents into natively editable PowerPoint decks Live demo ↗ (opens in a new tab)62.1k starsMIT93 out of 100
2browser-usePython library that lets an LLM agent control a web browser Live demo ↗ (opens in a new tab)118k starsMIT89 out of 100
3langchainPython framework for building LLM agents and apps from interoperable components147.8k starsMIT88 out of 100
#4tradingagentsLangGraph multi-agent framework simulating a trading firm to analyze tickers Live demo ↗ (opens in a new tab)111.2k starsApache-2.088 out of 100
#5copilotkitSDK for adding agent chat, generative UI and shared state to apps38k starsMIT87 out of 100