SGLang is a fast serving framework for large language models and vision language models.

Reverse Instructions to generate instruction tuning data with corpus examples

A collection of awesome-prompt-datasets, awesome-instruction-dataset, to train ChatLLM such as chatgpt 收录各种各样的指令数据集, 用于训练 ChatLLM 模型。

Entropy Based Sampling and Parallel CoT Decoding

Freeing data processing from scripting madness by providing a set of platform-agnostic customizable pipeline processing blocks.

Code for Paper: Training Software Engineering Agents and Verifiers with SWE-Gym [ICML 2025]

Stop Overthinking: A Survey on Efficient Reasoning for Large Language Models

Paper list for Efficient Reasoning.

Curated list of datasets and tools for post-training.

Awesome Reasoning LLM Tutorial/Survey/Guide

✨✨Latest Papers and Benchmarks in Reasoning with Foundation Models

Unified Efficient Fine-Tuning of 100+ LLMs & VLMs (ACL 2024)

An evaluation benchmark for undergraduate competition math in Lean4, Isabelle, Coq, and natural language.

A lightweight library for generating synthetic instruction tuning datasets for your data without GPT.

Code for "SemDeDup", a simple method for identifying and removing semantic duplicates from a dataset (data pairs which are semantically similar, but not exactly identical).

A reading list on LLM based Synthetic Data Generation 🔥

Code for the paper "Evaluating Large Language Models Trained on Code"

Awesome RL Reasoning Recipes ("Triple R")

Synthetic data curation for post-training and structured data extraction

[NeurIPS 2023 D&B] Code repository for InterCode benchmark https://arxiv.org/abs/2306.14898

The code and data for "MMLU-Pro: A More Robust and Challenging Multi-Task Language Understanding Benchmark" [NeurIPS 2024]

Automatic evals for LLMs

LIMO: Less is More for Reasoning

A streamlined and customizable framework for efficient large model evaluation and performance benchmarking

Open-source evaluation toolkit of large multi-modality models (LMMs), support 220+ LMMs, 80+ benchmarks