Official Repo for Open-Reasoner-Zero

A lightweight reproduction of DeepSeek-R1-Zero with indepth analysis of self-reflection behavior.

Fully open reproduction of DeepSeek-R1

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

g1: Using Llama-3.1 70b on Groq to create o1-like reasoning chains

🍼 Official implementation of Dynamic Data Mixing Maximizes Instruction Tuning for Mixture-of-Experts

A curated reading list of research in Mixture-of-Experts(MoE).

Arena-Hard-Auto: An automatic LLM benchmark.

Official repository for ICLR 2025 paper "Alignment Data Synthesis from Scratch by Prompting Aligned LLMs with Nothing". Your efficient and high-quality synthetic data generation pipeline!

[NeurIPS'24] Official code for *🎯DART-Math: Difficulty-Aware Rejection Tuning for Mathematical Problem-Solving*

Official repo for the paper "Scaling Synthetic Data Creation with 1,000,000,000 Personas"

Open source project for data preparation of LLM application builders

🗺️ Data Cleaning and Textual Data Visualization 🗺️

Distilabel is a framework for synthetic data and AI feedback for engineers who need fast, reliable and scalable pipelines based on verified research papers.

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

Scalable toolkit for efficient model alignment

👨‍💻 An awesome and curated list of best code-LLM for research.

Curated list of datasets and tools for post-training.

A bagel, with everything.

Reformatted Alignment

Generate multi-round conversation roleplay data based on self-instruct and evol-instruct.

PyTorch implementation of the paper "SuperLoss: A Generic Loss for Robust Curriculum Learning" in NIPS 2020.

A family of open-sourced Mixture-of-Experts (MoE) Large Language Models

Continual Learning of Large Language Models: A Comprehensive Survey

⛷️ LLaMA-MoE: Building Mixture-of-Experts from LLaMA with Continual Pre-training (EMNLP 2024)