Stars
very good whiteboard SDK / infinite canvas SDK
AgentScope: Agent-Oriented Programming for Building LLM Applications
💫 Toolkit to help you get started with Spec-Driven Development
About Awesome things towards foundation agents. Papers / Repos / Blogs / ...
No fortress, purely open ground. OpenManus is Coming.
An opinionated list of awesome Python frameworks, libraries, software and resources.
AI Analytics and Knowledge Engine for RAG over large-scale, heterogeneous data. - The only MCP Server you'll ever need
Fully open data curation for reasoning models
Use PEFT or Full-parameter to CPT/SFT/DPO/GRPO 500+ LLMs (Qwen3, Qwen3-MoE, Llama4, GLM4.5, InternLM3, DeepSeek-R1, ...) and 200+ MLLMs (Qwen3-VL, Qwen3-Omni, InternVL3.5, Ovis2.5, Llava, GLM4v, Ph…
Fast, Flexible and Portable Structured Generation
A high-throughput and memory-efficient inference and serving engine for LLMs
Production-ready platform for agentic workflow development.
Open Source Deep Research Alternative to Reason and Search on Private Data. Written in Python.
🤗 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.
A lightweight, powerful framework for multi-agent workflows
《开源大模型食用指南》针对中国宝宝量身打造的基于Linux环境快速微调（全参数/Lora）、部署国内外开源大模型（LLM）/多模态大模型（MLLM）教程
Curated list of datasets and tools for post-training.
LLMs interview notes and answers:该仓库主要记录大模型（LLMs）算法工程师相关的面试题和参考答案
Fully open reproduction of DeepSeek-R1
Building DeepSeek R1 from Scratch
Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.
PIKE-RAG: sPecIalized KnowledgE and Rationale Augmented Generation
OpenCompass is an LLM evaluation platform, supporting a wide range of models (Llama3, Mistral, InternLM2,GPT-4,LLaMa2, Qwen,GLM, Claude, etc) over 100+ datasets.
[EMNLP'23, ACL'24] To speed up LLMs' inference and enhance LLM's perceive of key information, compress the prompt and KV-Cache, which achieves up to 20x compression with minimal performance loss.
Chat with your database or your datalake (SQL, CSV, parquet). PandasAI makes data analysis conversational using LLMs and RAG.
DSPy: The framework for programming—not prompting—language models