Run the full 2.78-trillion-parameter Kimi K3 model beyond available RAM by streaming activated weights directly from NVMe. A dependency-free, embeddable C inference engine.

An explorable Japanese suburban railway-crossing neighbourhood on a small planet, rendered 3D-to-2D as a cel-shaded anime background. Three.js, no image assets.

Pebble

⌥ AI Coding agent for the terminal — hash-anchored edits, optimized tool harness, LSP, Python, browser, subagents, and more

Wrap Antigravity, ChatGPT Codex, Claude Code, Grok Build as an OpenAI/Gemini/Claude/Codex compatible API service, allowing you to enjoy the free Gemini 3.1 Pro, GPT 5.6 Series, Grok 4.5, Claude mod…

harnessing all the naturally occurring proof engineering data as evals | PROTOTYPING we haven't invested in the finalization/comms yet

Same model, different wrapper: a from-scratch benchmark comparing coding-agent harnesses (codex, pi, opencode, cursor, devin) and open models on correctness, speed, and token cost

A simple stateless privacy-protecting URL redirecting service for Matrix

Turn your PC, Mac, or Linux box into an AI server. LLM inference, chat UI, voice, agents, workflows, RAG, and image generation.

Real-time global intelligence dashboard. AI-powered news aggregation, geopolitical monitoring, and infrastructure tracking in a unified situational awareness interface

A skill to stop your coding agent from burying the answer. ADHD-friendly output.

BrainStem is a biologically inspired, Real Neuro-Symbolic (RNS-AI) cognitive architecture for lifelong learning. It is designed to learn models of the structures and dynamics of language and text t…

Mooncake is the serving platform for Kimi, a leading LLM service provided by Moonshot AI.

A Flexible Framework for Experiencing Heterogeneous LLM Inference/Fine-tune Optimizations

Skills for Real Engineers. Straight from my .claude directory.

LIBERATED AI CHAT

A minimal, distraction-free markdown editor built with Tauri, React, and TypeScript

A vector index built on TurboQuant, written in Rust with Python bindings

A vLLM patch + hand‑written SM120 SASS kernels: 2‑bit MoE experts + an FP4 "delta" cache that recovers precision — matching the official (NV)FP4 checkpoint's quality on consumer Blackwell cards

The Fable Workflow: how Claude Fable 5 worked, distilled into skills any model can run, with the eval that keeps it honest. Think / act / prove.

See what your Orca agents are doing — the task DAG, who is working what, and the gates blocking the run.

Run frontier MoE models on hardware you already own — pure C, zero deps, experts streamed from disk. Tiny engine, immense model. 🐦

Mixed-capability LLM benchmark for DGX Spark — 57 scenarios, 10 domains, partial-credit grading, trial statistics

Train a tiny Transformer from scratch on your iMessage history, entirely on your Mac.

Does relational-process register (Gowith) change a model's epistemic self-monitoring — and why? A mechinterp experiment: 6 register conditions × 3 binary-scored tasks on Gemma-3-12B-it + Gemma Scop…

Iris is a desktop voice companion that uses Gemini Live for natural realtime conversation and Hermes Agent for long-running work.

Building AI agents, atomically

Debug, evaluate, and monitor your LLM applications, RAG systems, and agentic workflows with comprehensive tracing, automated evaluations, and production-ready dashboards.