Stars
Rich is a Python library for rich text and beautiful formatting in the terminal.
Draft-Target Disaggregation LLM Serving System via Parallel Speculative Decoding.
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…
Dingo: A Comprehensive AI Data Quality Evaluation Tool
gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI
Ultra-high-performance, secure, all-in-one acceleration engine for developer resources whose performance far surpasses traditional accelerators, delivering a unified, efficient acceleration experie…
An AI agent development platform with all-in-one visual tools, simplifying agent creation, debugging, and deployment like never before. Coze your way to AI Agent creation.
Train speculative decoding models effortlessly and port them smoothly to SGLang serving.
[Lumina Embodied AI] 具身智能技术指南 Embodied-AI-Guide
Parallel Scaling Law for Language Model — Beyond Parameter and Inference Time Scaling
Open-source implementation of AlphaEvolve
Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.
🚀 Truly open-source AI avatar(digital human) toolkit for offline video generation and digital human cloning.
Lightning-Fast RL for LLM Reasoning and Agents. Made Simple & Flexible.
My learning notes/codes for ML SYS.
A Datacenter Scale Distributed Inference Serving Framework
VirtualWife是一个虚拟数字人项目，支持B站直播，支持openai、ollama
FlashMLA: Efficient Multi-head Latent Attention Kernels
MoBA: Mixture of Block Attention for Long-Context LLMs
A Flexible Framework for Experiencing Cutting-edge LLM Inference Optimizations
User-friendly AI Interface (Supports Ollama, OpenAI API, ...)
Fully open reproduction of DeepSeek-R1
基于大模型搭建的聊天机器人，同时支持 微信公众号、企业微信应用、飞书、钉钉 等接入，可选择ChatGPT/Claude/DeepSeek/文心一言/讯飞星火/通义千问/ Gemini/GLM-4/Kimi/LinkAI，能处理文本、语音和图片，访问操作系统和互联网，支持基于自有知识库进行定制企业智能客服。