#01Framework
ECC
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
272.5K stars93/100 healthJavaScript
Curated from the live project index
Compare 30 leading open source LLM runtimes, serving engines, interfaces, frameworks, and developer tools. The shortlist combines GitHub adoption with current momentum, maintenance health, and recent repository activity.
Side-by-side signals
| Project | Primary role | Stars | Weekly | Health | Language | Updated |
|---|---|---|---|---|---|---|
| ECC | Framework | 272.5K | +52.6K | Strong93/100 | JavaScript | 2026-10-02 02:01 UTC |
| hermes-agent | Framework | 251.1K | +74.1K | Strong95/100 | Python | 2026-10-04 09:18 UTC |
| firecrawl | LLM tooling | 188.4K | +36.0K | Strong80/100 | TypeScript | 2026-10-03 18:46 UTC |
| AutoGPT | LLM tooling | 187.6K | +174 | Strong88/100 | Python | 2026-10-04 01:21 UTC |
| ollama | Runtime & serving | 182.2K | +450 | Strong88/100 | Go | 2026-10-04 05:57 UTC |
| prompts.chat | LLM tooling | 172.0K | +4.2K | Healthy77/100 | HTML | 2026-10-03 08:12 UTC |
| transformers | Runtime & serving | 166.9K | +220 | Strong89/100 | Python | 2026-10-04 01:58 UTC |
| dify | Framework | 157.8K | +815 | Strong92/100 | TypeScript | 2026-10-03 15:55 UTC |
| langflow | Interface | 155.5K | +4.2K | Strong85/100 | Python | 2026-10-04 08:06 UTC |
| ponytail | Framework | 154.0K | +2.3K | Strong84/100 | JavaScript | 2026-10-03 05:12 UTC |
| open-webui | Runtime & serving | 153.9K | +1.0K | Strong93/100 | Python | 2026-10-03 13:06 UTC |
| langchain | Framework | 147.4K | +599 | Strong84/100 | Python | 2026-10-03 08:31 UTC |
| awesome-llm-apps | Framework | 140.7K | +16.4K | Healthy76/100 | Python | 2026-09-30 20:31 UTC |
| llama.cpp | Runtime & serving | 130.3K | +2.9K | Strong98/100 | C++ | 2026-10-04 08:54 UTC |
| MoneyPrinterTurbo | LLM tooling | 128.3K | +847 | Strong86/100 | Python | 2026-10-04 06:48 UTC |
Explore the shortlist
The agent harness performance optimization system. Skills, instincts, memory, security, and research-first development for Claude Code, Codex, Opencode, Cursor and beyond.
The agent that grows with you
Supercharge your AI agents with data from the web and beyond. Building the library for superintelligence. 🔥
AutoGPT is the vision of accessible AI for everyone, to use and to build on. Our mission is to provide the tools, so that you can focus on what matters.
Get up and running with Kimi, GLM, MiniMax, DeepSeek, gpt-oss, Qwen, Gemma and other models.
f.k.a. Awesome ChatGPT Prompts. Share, discover, and collect prompts from the community. Free and open source — self-host for your organization with complete privacy.
🤗 Transformers: the model-definition framework for state-of-the-art machine learning models in text, vision, audio, and multimodal models, for both inference and training.
Build Agentic workflows, RAG pipelines, with rich AI model and tool support on one collaborative workspace. Deploy on cloud, VPC, or self-hosted, so teams move from prototype to production without rebuilding the stack.
Langflow is a powerful tool for building and deploying AI-powered agents and workflows.
Makes your AI agent think like the laziest senior dev in the room. The best code is the code you never wrote.
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
The agent engineering platform.
100+ AI Agents, Agent Skills and RAG Apps - Free and Open Source.
LLM inference in C/C++
利用 AI 大模型和自动化工作流，根据主题或关键词一键生成高清短视频。Generate HD short videos from a topic or keyword with an automated AI workflow.
Turn any codebase, with its docs, SQL schemas, configs, and PDFs, into a queryable knowledge graph. A /graphify skill for Claude Code, Cursor, Codex, and Gemini CLI: local deterministic AST parsing, every edge explained, no vector store.
21 Lessons, Get Started Building with Generative AI
Agents that use the browser.
TradingAgents: Multi-Agents LLM Financial Trading Framework
🪨 why use many token when few token do trick. Viral skill + proxy for coding agents that cuts 65% of tokens by talking like a caveman.
Implement a ChatGPT-like LLM in PyTorch from scratch, step by step
A high-throughput and memory-efficient inference and serving engine for LLMs
RAGFlow is a leading open-source Retrieval-Augmented Generation (RAG) engine that fuses cutting-edge RAG with Agent capabilities to create a superior context layer for LLMs
Turn any PDF or image document into structured data for your AI. A powerful, lightweight OCR toolkit that bridges the gap between images/PDFs and LLMs. Supports 100+ languages.
Open-source web crawler and scraper for LLMs and AI agents: any website into clean, LLM-ready Markdown. Run it yourself, or use Crawl4AI Cloud with one key.
Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.
🐙 Guides, papers, lessons, notebooks and resources for prompt engineering, context engineering, RAG, and AI Agents.
An open-source long-horizon SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skill, subagents and message gateway, it handles different levels of tasks that could take minutes to hours.
Turn any idea, plan, or codebase into a beautiful interactive diagram. An agent skill for Claude Code, Codex, and more.
AI agent toolkit: unified LLM API, agent loop, TUI, coding agent CLI
How to choose
Prioritize hardware support, quantization formats, setup friction, model compatibility, and offline operation.
Evaluate throughput, latency, batching, observability, scaling, API compatibility, and accelerator support.
Check extension points, data connectors, authentication, deployment model, user experience, and upgrade cadence.
Quick answers
This page includes infrastructure and applications built around large language models: local runtimes, inference and serving engines, developer frameworks, model interfaces, evaluation tools, and other repositories with a clear LLM signal. It is not limited to model-weight repositories.
Choose by the layer you need. A local runtime is useful for running models on a workstation, a serving engine targets production inference, an interface adds an end-user experience, and a framework helps developers build applications. Compare projects within the same role before choosing.
Stars are useful evidence of awareness and adoption, but they are not a quality guarantee. Maintenance health, recent releases, hardware support, license, throughput, integrations, and your deployment constraints can be more important.
It depends on both the software license and the license of the model you run. Review the upstream repository, model terms, dependencies, and hosted-service restrictions before commercial use.
Get the fastest-growing projects, useful MCP servers, and technical reads in one weekly email.