a benchmark to evaluate the situated inductive reasoning

Learn online intrinsic rewards from LLM feedback

Sandboxed code execution for AI agents, locally or on the cloud.

A lightweight task engine for building stateful AI agents that prioritizes simplicity and flexibility.

Benchmarking Agentic LLM and VLM Reasoning On Games

RAG that intelligently adapts to your use case, data, and queries

The first AI agent that builds third-party integrations through reverse engineering platforms' internal APIs.

Routing on Random Forest (RoRF)

AWM: Agent Workflow Memory

Synthetic data derived by templating, few shot prompting, transformations on public domain corpora, and monte carlo tree search.

home of hitorilabs

Long context evaluation for large language models

Your first AI prompt engineer

Code for Quiet-STaR

Code and data used in the paper: "Training on Incorrect Synthetic Data via RL Scales LLM Math Reasoning Eight-Fold"

Playing Pokemon Red with Reinforcement Learning

Open-source framework for exporting your personal data.

🐚 OpenDevin: Code Less, Make More

The AI Scientist: Towards Fully Automated Open-Ended Scientific Discovery 🧑‍🔬

Tree Attention: Topology-aware Decoding for Long-Context Attention on GPU clusters

This is the repository that contains the source code for the Self-Evaluation Guided MCTS for online DPO.

LLMs + Lean, on your laptop or in the cloud

Code for paper "Interpretability in Action: Exploratory Analysis of VPT, a Minecraft Agent"

A comprehensive repository of reasoning tasks for LLMs (and beyond)

🏛️A research-friendly codebase for fast experimentation of single-agent reinforcement learning in JAX • End-to-End JAX RL