Toward High-Accuracy Open-Source Biomolecular Structure Prediction.

LLM RL envs done right, plus some training code

Run OpenClaw, (formerly Moltbot, formerly Clawdbot) on Cloudflare Workers

Compression for unit-norm embedding vectors using spherical coordinates

The open source coding agent.

Complete Claude Code configuration collection - agents, skills, hooks, commands, rules, MCPs. Battle-tested configs from an Anthropic hackathon winner.

Production-ready platform for agentic workflow development.

Accomplish™ (formerly Openwork) is the open source Al coworker that lives on your desktop

🐹 Deep clean and optimize your Mac.

PDLP algorithm for linear programming

The similarity score for spectral comparison

A unified inference and post-training framework for accelerated video generation.

Official repository of PXDesign

AgentEvolver: Towards Efficient Self-Evolving Agent System

[ICLR 2025] When Attention Sink Emerges in Language Models: An Empirical View (Spotlight)

Puzzles for learning Triton, play it with minimal environment configuration!

Perplexity open source garden for inference technology

Generate cuda kernels for linear attention mechanisms like RWKV and Mamba

PDF pipeline for creating training corpora (mainly for llm, multimodal and alignment horizontals)

Ling-V2 is a MoE LLM provided and open-sourced by InclusionAI.

Training framework with a goal to explore the frontier of sample efficiency of small language models

slime is an LLM post-training framework for RL Scaling.

Contexts Optical Compression

Implementation of "Hyperspherical Latents Improve Continuous-Token Autoregressive Generation"

Democratizing Reinforcement Learning for LLMs

🍒 Cherry Studio is a desktop client that supports for multiple LLM providers.