Train the smallest LM you can that fits in 16MB. Best model wins!
Repositories
Hmbown repositories
Orchestrate coding agents remotely from your phone, desktop and CLI
Logic architecture for abductive inference. Forces LLMs to make uncertainty visible through explicit hypothesis generation and Inference to Best Explanation.
Terminal pixel-art office for AI coding agents
Prime Agent is a self-improving RLM agent that can be used for coding workflows, or even long-running autonomous tasks.
CUDA/Triton kernel optimization and runtime patching for LLM inference
Registry of agents implementing the Agent Client Protocol (ACP)
General plug-and-play inference library for Recursive Language Models (RLMs), supporting various sandboxes.
RLM agent harness - built on Deep Agents
Superlinked Inference Engine is an Open-source inference server and production cluster for embeddings, reranking, and extraction.
Public repository for Agent Skills
slime is an LLM post-training framework for RL Scaling.
Symphony turns project work into isolated, autonomous implementation runs, allowing teams to manage work instead of supervising coding agents.
DSPy reasoning harness for four-corner (P, not-P, both, neither) analysis of hard questions
You like pytorch? You like micrograd? You love tinygrad! ❤️
A high-throughput and memory-efficient inference and serving engine for LLMs
OpenWarp is a free version of the open source client based on warp
MCP server for recursive market analysis. Yahoo Finance + CoinGecko + sandboxed Python + RLM sub-queries via claude/codex/gemini/copilot CLI. Zero API keys required.