Stars
🚧 Accepting Task Submissions 🚧
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…
Flash Attention Triton kernel with support for second-order derivatives
Teams-first Multi-agent orchestration for Claude Code
Kernel sources for https://huggingface.co/kernels-community
GitHub is for humans. AgentHub is for agents. First use case is for autoresearch but it's a lot more general than that. Exploratory project.
Train the smallest LM you can that fits in 16MB. Best model wins!
Run OpenClaw more securely inside NVIDIA OpenShell with managed inference
Generate embeddings from large-scale graph-structured data.
Official implementation of RAE-NWM: Navigation World Model in Dense Visual Representation Space.
Official Codebase for "Neural Thickets: Diverse Task Experts Are Dense Around Pretrained Weights"
AI agents running research on single-GPU nanochat training automatically
Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents.
Code and website for Self-Flow: Self-Supervised Flow Matching for Scalable Multi-Modal Synthesis
The Automated LLM Speedrunning Benchmark measures how well LLM agents can reproduce previous innovations and discover new ones in language modeling.
Single-step image generation at 306 FPS. Drifting vs Diffusion head-to-head on CIFAR-10.
Midway Network: Learning Representations for Recognition and Motion from Latent Dynamics
Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞
OpenVision (ICCV 2025), OpenVision 2 (CVPR 2026), and OpenVision 3
Scaling Text-to-Image Diffusion Transformers with Representation Autoencoders
Conditional Memory via Scalable Lookup: A New Axis of Sparsity for Large Language Models
Algorithm powering the For You feed on X
A compact implementation of SGLang, designed to demystify the complexities of modern LLM serving systems.
Pixio: a capable vision encoder dedicated to dense prediction, simply by pixel reconstruction
Code for Bolmo: Byteifying the Next Generation of Language Models