Best Open Source LLMs

Langflow is a powerful tool for building and deploying AI-powered agents and workflows.

Implement a ChatGPT-like LLM in PyTorch from scratch, step by step

Course to get into Large Language Models (LLMs) with roadmaps and Colab notebooks.

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

为GPT/GLM等LLM大语言模型提供实用化交互接口，特别优化论文阅读/润色/写作体验，模块化设计，支持自定义快捷按钮&函数插件，支持Python和C++等项目剖析&自译解功能，PDF/LaTex论文翻译&总结功能，支持并行问询多种LLM模型，支持chatglm3等本地模型。接入通义千问, deepseekcoder, 讯飞星火, 文心一言, llama2, rwkv, claude2, moss等。

Ultra-lightweight, open-source, self-hosted personal AI agent framework in Python with WebUI, tools, memory, MCP, multi-agent workflows, automation, and chat apps

《深入理解 AI Agent：设计原理与工程实践》（李博杰 著）开源主仓库：全书正文、编译版 PDF 与按章配套代码

Ray is an AI compute engine. Ray consists of a core distributed runtime and a set of AI Libraries for accelerating ML workloads.

[EMNLP2025] LightRAG: Simple and Fast Retrieval-Augmented Generation

DeepTutor: Lifelong Personalized Tutoring. https://deeptutor.info/.

A Python library for extracting structured information from unstructured text using LLMs with precise source grounding and interactive visualization.

🪢 Open source agent evals & observability: Trace, evaluate, and improve LLM applications with one open platform.

Build and run agents you can see, understand and trust.

An LLM-powered knowledge curation system that researches a topic and generates a full-length report with citations.

A one stop repository for generative AI research updates, interview resources, notebooks and much more!

Official code repo for the O'Reilly Book - "Hands-On Large Language Models"

Open-source AI orchestration framework for building context-engineered, production-ready LLM applications. Design modular pipelines and agent workflows with explicit control over retrieval, routing, memory, and generation. Built for scalable agents, RAG, multimodal applications, semantic search, and conversational systems.

Distribute and run LLMs with a single file.

The official repo of Qwen (通义千问) chat & pretrained large language model proposed by Alibaba Cloud.

FinGPT: Open-Source Financial Large Language Models! Revolutionize 🔥 We release the trained model on HuggingFace.

Code for Machine Learning for Trading, 3rd edition - from data sourcing to live execution.

Machine Learning Engineering Open Book

中文LLaMA&Alpaca大语言模型+本地CPU/GPU训练部署 (Chinese LLaMA & Alpaca LLMs)

:sparkles::sparkles:Latest Advances on Multimodal Large Language Models

Ongoing research training transformer models at scale

🐫 CAMEL: The first and the best multi-agent framework. Finding the Scaling Law of Agents. https://www.camel-ai.org

Sample code and notebooks for Generative AI on Google Cloud, with Gemini Enterprise Agent Platform

Automated Penetration Testing Agentic Framework Powered by Large Language Models

20+ high-performance LLMs with recipes to pretrain, finetune and deploy at scale.

💡 All-in-one AI framework for semantic search, LLM orchestration and language model workflows

Pocket Flow: Codebase to Tutorial

A curated list of modern Generative Artificial Intelligence projects and services

Hierarchical Reasoning Model Official Release

An open-source, tool-augmented conversational language model from Fudan University

Pocket Flow: 100-line LLM framework. Let Agents build Agents!

A curated list of 120+ LLM libraries category wise.

A curated list of practical guide resources of LLMs (LLMs Tree, Examples, Papers)

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & VLM & TIS & vLLM & Ray & Async RL)

High-speed Large Language Model Serving for Local Deployment

A straightforward method for training your LLM, from downloading data to generating text.

Qwen3.8-27B on a single RTX 3090 with vLLM: ~1,000 tok/s at 64 concurrent (int8 tensor-core GEMMs, fp16 DeltaNet state), ~114 tok/s single-user at default sampling / ~124 greedy (MTP drafts, own-output draft vocab, calibrated int4 lm_head, split-KV verify attention), 150k-262k context; patches, requant scripts, benchmarks

Run full Kimi K3 on a single device. And an OpenAI-compatible API server for local chat and coding agents.

Open-source AI companions, desktop pets, long-term memory & proactive chat. 人机恋开源项目大全。让你的家机能够脱离官端自主存在，拥有记忆和主动性。

Up to 4× faster LLM decoding on Apple Silicon, lossless. Native MLX port of DeepSeek's DSpark & z-lab's DFlash speculative decoding - Gemma-4, Qwen3.8, Muse-Glimmer, Nemotron, LFM2.5, Ornith-1.0, ternary Bonsai-27B.

Swiftlet is a Swift and Metal runtime that runs large Qwen Mixture-of-Experts models locally on Apple devices by streaming expert weights from storage, enabling 35B and 80B models to run with low RAM, including on iPhone.

The fastest way to run Qwen3.8-Flash-Next on Strix Halo (gfx1151)

Run Qwen3.8-Flash-Next (125B MoE, 104 GB at 4-bit) on Macs with a fraction of that RAM by streaming experts from SSD. MLX + Swift, Ollama-compatible API.

NeoHorse-1: Towards Recursive Self-Improvement via Agentic Post-Training with Routing Harness.

An intentionally vulnerable OWASP LLM Top 10 training platform for AI Security, Prompt Injection, RAG Security, Agent Security, and GenAI penetration testing.

MLNLP社区用来帮助大家论文Rebuttal的整理仓库。

Your car as a chat-room agent: Raspberry Pi 5 + dashcam + local AI. CodeWatch's sibling for the garage.

Every device brings a slice. Together they run the whole model. Peer-to-peer LLM inference across browser tabs: a from-scratch WebGPU engine and a WebRTC runtime that split a 27B model over the devices in a room.

Learn LLM Inference Engineering step by step - from KV cache, PagedAttention, and continuous batching to vLLM, SGLang, and GPUs.

Your AI intranet: network the computers you already own for inference and training.

Jacobian-Brainwash : A manual alignment tool for large language models built on Anthropic's Jacobian Lens. Results are exportable.

从 MiniMind 源码读起,再延伸到现代大模型技术体系的中文学习笔记。主线逐行精读预训练 / SFT / DPO / PPO / GRPO 与训练机制;附录 17 篇进阶卷覆盖量化、投机解码、RLHF 全景、模型代际史等 MiniMind 没涉及、但进阶绕不开的主题。

Transformer architecture explained step by step - the full architecture, every attention variant, positional embeddings, and every layer inside a Transformer.

SSD-streaming inference engine for giant MoE models (Rust + CUDA). GLM 5.2 743B at 2 tok/s and Hy3 295B at 7 tok/s on two consumer 16GB GPUs. Zero-config multi-GPU: measures PCIe bandwidth, places attention and hot experts where they fit.

The free open source offline writing assistant assist with grammar, rewriting & tone entirely on your machine.

One CLI for all your robots. Connect them, command them, and let them work together, each with an LLM for a brain. Microduck, Open Duck Mini, LeRobot arms, XLeRobot carts, AlohaMini, ToddlerBot or any ROS base. Claude, OpenAI, Gemini, Grok, or local via Ollama or vLLM. Simulator, .duck safety contracts, MCP, memory between runs, flocks.