Write scalable load tests in plain Python 🚗💨

A streamlined and customizable framework for efficient large model evaluation and performance benchmarking

LLMPerf is a library for validating and benchmarking LLMs

The trust-minimized, zero-knowledge bridging protocol, designed for censorship resistance, extremely high security, and usage in decentralized finance.

✨ 易上手的多平台 LLM 聊天机器人及开发框架 ✨ 平台支持 QQ、QQ频道、Telegram、微信、企微、飞书 | OpenAI、DeepSeek、Gemini、硅基流动、月之暗面、Ollama、OneAPI、Dify 等。附带 WebUI。

Toolkit for linearizing PDFs for LLM datasets/training

The python library for real-time communication

A framework for few-shot evaluation of language models.

Implementation of python itertools and builtin iteration functions for C++17

WiFi密码暴力破解工具-图形界面，支持WPA/WPA2/WPA3、多开并发、自动破解、自定义密码本、自动生成密码字典

Optimized implementations of various library functions for ARM architecture processors

EasyR1: An Efficient, Scalable, Multi-Modality RL Training Framework based on veRL

A hub for various industry-specific schemas to be used with VLMs.

Animating R1's thoughts.

An open Apple AirDrop implementation written in Python

Materials for learning SGLang

[NeurIPS DB Track, 2024] MATH-Vision dataset and code to measure multimodal mathematical reasoning capabilities.

SpargeAttention: A training-free sparse attention that can accelerate any model inference.

Official codebase for "SWE-RL: Advancing LLM Reasoning via Reinforcement Learning on Open Software Evolution"

A lightweight data processing framework built on DuckDB and 3FS.

Dynamic Memory Management for Serving LLMs without PagedAttention

A high-performance distributed file system designed to address the challenges of AI training and inference workloads.

High-speed GEMV kernels, at most 2.7x speedup compared to pytorch baseline.

DeepGEMM: clean and efficient FP8 GEMM kernels with fine-grained scaling

Official implementation of Half-Quadratic Quantization (HQQ)

DeepEP: an efficient expert-parallel communication library

Claude Code is an agentic coding tool that lives in your terminal, understands your codebase, and helps you code faster by executing routine tasks, explaining complex code, and handling git workflo…

A CPU Realtime VLM in 500M. Surpassed Moondream2 and SmolVLM. Training from scratch with ease.

修正文档扭曲/模糊/阴影等情况，使用onnx模型简单轻量部署，未来持续跟进最新最好的文档矫正方案和模型,Correct document distortion using a lightweight ONNX model for easy deployment. We will continue to follow and integrate the latest and best docu…