A Python-embedded DSL that makes it easy to write fast, scalable ML kernels with minimal boilerplate.

Distributed Compiler based on Triton for Parallel Systems

Domain-specific language designed to streamline the development of high-performance GPU/CPU/Accelerators kernels

A PyTorch native platform for training generative AI models

CUDA Templates and Python DSLs for High-Performance Linear Algebra

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.

Memory optimization and training recipes to extrapolate language models' context length to 1 million tokens, with minimal hardware.

A curated list for Efficient Large Language Models

Fine-tuning & Reinforcement Learning for LLMs. 🦥 Train OpenAI gpt-oss, DeepSeek-R1, Qwen3, Gemma 3, TTS 2x faster with 70% less VRAM.

Tools for merging pretrained large language models.

🤗 A specialized library for integrating context-free grammars (CFG) in EBNF with the Hugging Face Transformers

Enforce the output format (JSON Schema, Regex etc) of a language model

Train transformer language models with reinforcement learning.

Utilities for Training Very Large Models

Code associated with the paper **Draft & Verify: Lossless Large Language Model Acceleration via Self-Speculative Decoding**

A Toolbox for Adversarial Robustness Research

Kernl lets you run PyTorch transformer models several times faster on GPU with a single line of code, and is designed to be easily hackable.

[ACL 2023] Code for ContraCLM: Contrastive Learning For Causal Language Model

Code and data for "Lost in the Middle: How Language Models Use Long Contexts"

Benchmarking large language models' complex reasoning ability with chain-of-thought prompting

LLMs build upon Evol Insturct: WizardLM, WizardCoder, WizardMath

A framework for the evaluation of autoregressive code generation language models.

Hackable and optimized Transformers building blocks, supporting a composable construction.

An easy-to-use LLMs quantization package with user-friendly apis, based on GPTQ algorithm.

Extend the original llama.cpp repo to support redpajama model.

ZJU RVPN 客户端的 Go 语言实现

Large Language Model Text Generation Inference

Open Instruction Generalist is an assistant trained on massive synthetic instructions to perform many millions of tasks