Official repository for the paper "LiveCodeBench: Holistic and Contamination Free Evaluation of Large Language Models for Code"

The Learning Interpretability Tool: Interactively analyze ML models to understand their behavior in an extensible and framework agnostic interface.

Harbor is a framework for running agent evaluations and creating and using RL environments.

Post-training with Tinker

An instruction-based benchmark for text improvements.

The 100 line AI agent that solves GitHub issues or helps you in your command line. Radically simple, no huge configs, no giant monorepo—but scores >74% on SWE-bench verified!

Fast and accurate AI powered file content types detection

Language Savant. If your repository's language is being reported incorrectly, send us a pull request!

General Reasoner: Advancing LLM Reasoning Across All Domains [NeurIPS25]

A faster file programming language detector

Data generation and training repository for SERA: Soft-Verified Efficient Repository Agents.

Specification and documentation for Agent Skills

Reinforcement Learning via Self-Distillation (SDPO)

Google Research

Renderer for the harmony response format to be used with gpt-oss

gpt-oss-120b and gpt-oss-20b are two open-weight language models by OpenAI

PyTorch native post-training library

`wc` for tokens, using HuggingFace Tokenizers in Rust

🤗 Evaluate: A library for easily evaluating machine learning models and datasets.

Task-based datasets, preprocessing, and evaluation for sequence models.

Website for "A Survey of Machine Learning for Big Code and Naturalness"

regular expression library

Original Implementation of Prompt Tuning from Lester, et al, 2021