Open Source LLM Projects
Discover 274 open source LLM repositories from GitHub, each with an analysis of what it does, key features, use cases and alternatives. LLM projects here are most often combined with Python, AI Agents and AI. Last updated October 4, 2026.
274 repositories · updated October 4, 2026
openai-agents-js: Build Multi-Agent and Voice Workflows
A TypeScript SDK for building LLM agents that use tools, delegate work, and manage conversations. It also supports sandboxed tasks and low-latency voice agents, making it useful for teams building more than a single prompt-and-response flow.
gptpdf: Convert PDFs into Markdown with Vision Models
gptpdf turns PDF pages into Markdown using a vision-capable model, aiming to preserve layouts such as tables, formulas, and figures. It suits developers who need an API-based PDF parsing component and can provide access to a compatible model.
Ditto: Generate Flask Apps from Natural Language
Ditto is an experimental coding agent that turns a plain-language app description into a multi-file Flask application. It suits developers exploring LLM-assisted scaffolding, but the project is a quick exploration rather than a maintained production tool.
openllmetry: Observe LLM Applications with OpenTelemetry
OpenLLMetry adds OpenTelemetry-based tracing and metrics for LLM applications, providers, frameworks, and vector databases. It suits teams that want AI-specific observability in their existing telemetry stack rather than a separate monitoring format.
OLMoE.swift: Run OLMoE Locally on Apple Devices
OLMoE.swift is an Apple app for chatting with Ai2’s OLMoE language model on-device. It suits people who want private, offline-capable model use on iOS or macOS, provided their device can run the app and model.
SWE-agent: Use Language Models to Fix GitHub Issues
SWE-agent gives a language model tools to work on real software repositories, including diagnosing and attempting to fix GitHub issues. It is designed for software engineering research and configurable coding tasks, though its maintainers now recommend mini-SWE-agent for new use.
harmony: Render and Parse OpenAI Harmony Conversations
OpenAI Harmony is a Rust-based library with Python bindings for rendering and parsing the conversation format used by gpt-oss. It is useful for developers building inference systems that need to handle Harmony messages, channels, and tool calls directly.
plexe: Build Machine Learning Models from Prompts
Plexe turns a natural-language description and tabular dataset into a trained, packaged machine learning model through a multi-agent workflow. It is aimed at developers and data teams who want to automate model selection and iteration while retaining control over configuration and deployment.
Operit: Run AI Agents and Chat on Android
Operit is an Android AI agent platform that connects cloud or local models to device tools, terminals, browsers, and project workspaces. It suits Android users who want AI-assisted automation, development, and chat in one app.
instructor: Extract Validated Structured Data from LLMs
Instructor is a Python library that turns LLM responses into validated, typed data using Pydantic models. It suits applications that need reliable extraction across supported model providers without writing custom parsing and validation flows.
ChatTTS: Generate Expressive Speech for Dialogue
ChatTTS is a generative text-to-speech model built for dialogue, including assistant-style speech. It supports Chinese and English, multiple speakers, and controls for prosody such as pauses and laughter.
claude-code-proxy: Connect Claude Code to OpenAI and Gemini
A Python proxy lets Anthropic-compatible clients such as Claude Code use OpenAI, Gemini, or direct Anthropic backends. It translates requests and responses through LiteLLM, with configurable model mappings and provider selection.