Hy3 (295B A21B), a leading reasoning and agent model in its size, with great cost efficiency.

DFlash: Block Diffusion for Flash Speculative Decoding

Bleeding-edge ComfyUI for NVIDIA DGX Spark (GB10/Blackwell/sm_121a). CUDA 13 + SageAttention v3 (sm_121a) + NVFP4 + 14 custom-node packs + Flux 2 Dev / LTX 2.3 22B / ACE-Step v1.5 XL Turbo pre-bund…

High performance self-hosted photo and video management solution.

An interactive NVIDIA-GPU process viewer and beyond, the one-stop solution for GPU process management.

Own your compute, own your intelligence. Time to build your Personal AI Data Center.

Dont guess. Simulate. Multi-agent market prediction engine.

Use Codex from Claude Code to review code or delegate tasks.

MOVED to github.com/professorpalmer/marionette -- this repo is archived. Marionette: a desktop AI coding harness over Puppetmaster durable state.

A unified library of SOTA model optimization techniques like quantization, distillation, pruning, neural architecture search, speculative decoding, etc. It compresses deep learning models for downs…

Zenith: a continuous-improvement harness for long-running agent tasks. Turns Claude Code, Codex, or Hermes into a multi-agent mission orchestrator via MCP/ACP.

graff — a fast agentic coding harness in Zig: multi-provider, MCP, workflows, DGM evolution loop, TS/Python SDKs

AirPods liberated from Apple's ecosystem.

Self-contained DeepSeek V4 Flash DSpark TP=2 recipe for 2x DGX Spark with 62 tok/s benchmark

DeepSpec: a full-stack codebase for training and evaluating speculative decoding algorithms

Go ahead and axolotl questions

GLM-5: From Vibe Coding to Agentic Engineering

Programmable spreadsheet platform with a TypeScript/Rust engine, SDK, and embeddable workbook runtime.

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

cuTile Rust provides a safe, tile-based kernel programming DSL for the Rust programming language. It features a safe host-side API for passing tensors to asynchronously executed kernel functions.

The new generation chameleon based on NRF52840 makes the performance of card emulation more stable. And gave the chameleon the ability to read, write, and decrypt cards.

Tiny Model, Big Logic: Diversity-Driven Optimization Elicits Large-Model Reasoning Ability in VibeThinker-1.5B

Optimize prompts, code, and more with AI-powered Reflective Optimization

The agent that grows with you

Measuring frontier coding agents on original, long-horizon engineering tasks

Pier is a Harbor fork built for DeepSWE, with stronger support for CLI agents in air-gapped (no-internet) tasks and more faithful, consistent agent trajectories

The AI coding agent that runs on stolen Chipotle compute 🌯 Fork of OpenCode with Pepper AI as default model. Community project to add providers from Home Depot, Lowes, Target, Starbucks & more.

MLX-VLM is a package for inference and fine-tuning of Vision Language Models (VLMs) on your Mac using MLX.

Self-hosted AI workspace.

A Datacenter Scale Distributed Inference Serving Framework