locussam

anything-llm

The all-in-one Desktop & Docker AI application with built-in RAG, AI agents, and more.

firecrawl

🔥 Turn entire websites into LLM-ready markdown or structured data. Scrape, crawl and extract with a single API.

triton

Development repository for the Triton language and compiler

RWKV-LM

RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.

text-generation-inference

Large Language Model Text Generation Inference

Yi

A series of large language models trained from scratch by developers @01-ai

anthropic-cookbook

A collection of notebooks/recipes showcasing some fun and effective ways of using Claude.

courses

Anthropic's educational courses

lmdeploy

LMDeploy is a toolkit for compressing, deploying, and serving LLMs.

Book-Mathematical-Foundation-of-Reinforcement-Learning

This is the homepage of a new book entitled "Mathematical Foundations of Reinforcement Learning."

DeepSeek-V2

DeepSeek-V2: A Strong, Economical, and Efficient Mixture-of-Experts Language Model

Liger-Kernel

Efficient Triton Kernels for LLM Training

lectures

Material for cuda-mode lectures

darkriscv

opensouce RISC-V cpu core implemented in Verilog from scratch in one night!

rwkv.cpp

INT4/INT5/INT8 and FP16 inference on CPU for RWKV language model

vortex

Mooncake

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

claude-prompt-generator

Awesome-LLMs-on-device

Awesome LLMs on Device: A Comprehensive Survey

AlphaFold3

Open source implementation of AlphaFold3

Nanoflow

A throughput-oriented high-performance serving framework for LLMs

rtp-llm

RTP-LLM: Alibaba's high-performance LLM inference engine for diverse applications.

Yi-Coder

🌟 Yi-Coder is a series of open-source code language models that delivers state-of-the-art coding performance with fewer than 10 billion parameters.

hai-platform

一种任务级GPU算力分时调度的高性能深度学习训练平台

fast-voice-assistant

⚡ Insanely fast AI voice assistant with <500ms response times

sarathi-serve

A low-latency & high-throughput serving engine for LLMs

ml-slowfast-llava

SlowFast-LLaVA: A Strong Training-Free Baseline for Video Large Language Models

LongRoPE

Implementation of the LongRoPE: Extending LLM Context Window Beyond 2 Million Tokens Paper

TEAL

MagicDec

Breaking Throughput-Latency Trade-off for Long Sequences with Speculative Decoding