Train the smallest LM you can that fits in 16MB. Best model wins!

Benchmarking approaches to fine-tune AlphaGenome on lentiMPRA data

AI agents running research on single-GPU nanochat training automatically

⚒ Evolutionary self-improvement for Hermes Agent — optimize skills, prompts, and code using DSPy + GEPA

AI agents running research on single-GPU nanochat training automatically

Pixel office.

An open-source SuperAgent harness that researches, codes, and creates. With the help of sandboxes, memories, tools, skills and subagents, it handles different levels of tasks that could take minute…

Ralph is an autonomous AI agent loop that runs repeatedly until all PRD items are complete.

Train transformer language models with reinforcement learning.

An Easy-to-use, Scalable and High-performance Agentic RL Framework based on Ray (PPO & DAPO & REINFORCE++ & TIS & vLLM & Ray & Async RL)

AllenAI's post-training codebase

Search-R1: An Efficient, Scalable RL Training Framework for Reasoning & Search Engine Calling interleaved LLM based on veRL

Replication archive for "Do Claude Code and Codex P-Hack? Sycophancy and Statistical Analysis in Large Language Models"

OpenSandbox is a general-purpose sandbox platform for AI applications, offering multi-language SDKs, unified sandbox APIs, and Docker/Kubernetes runtimes for scenarios like Coding Agents, GUI Agent…

Official PyTorch Implementation for Learning a Generative Meta-Model of LLM Activations

AI agent toolkit: coding agent CLI, unified LLM API, TUI & web UI libraries, Slack bot, vLLM pods

Hypernetworks that update LLMs to remember factual information

Scaling Preference Data Curation via Human-AI Synergy

Single File, Single GPU, From Scratch, Efficient, Full Parameter Tuning library for "RL for LLMs"

HumanLM: Simulating Users with State Alignment Beats Response Imitation

Reading notes about Multimodal Large Language Models, Large Language Models, and Diffusion Models

Code and Data for Tau-Bench

An ARC-AGI solution using Agentica from Symbolica

"🐈 nanobot: The Ultra-Lightweight OpenClaw"

pip install continualcode

Official MCP server implementation for accessing Open Targets Data

Ideas for projects related to Tinker

Reinforcement Learning via Self-Distillation (SDPO)

Your own personal AI assistant. Any OS. Any Platform. The lobster way. 🦞