MiniMax-AI/SynLogic

[NeurIPS 2025] The official repo of SynLogic: Synthesizing Verifiable Reasoning Data at Scale for Learning Logical Reasoning and Beyond

42
/ 100
Emerging

SynLogic helps AI researchers and developers create high-quality, verifiable training data to improve the logical reasoning abilities of large language models (LLMs). It takes parameters for various logical tasks like Sudoku or Arrow Maze and generates an unlimited amount of diverse problem-solution pairs, which can then be used to train LLMs. This is for AI practitioners focused on developing and enhancing LLMs to perform complex reasoning.

198 stars. No commits in the last 6 months.

Use this if you need to generate large, customizable datasets of logical reasoning problems and their verifiable solutions to train or fine-tune large language models.

Not ideal if you are looking for an out-of-the-box LLM to solve real-world problems directly without involving model training or data generation.

AI model training LLM development Reasoning data synthesis Reinforcement learning from AI feedback Machine learning research
Stale 6m No Package No Dependents
Maintenance 2 / 25
Adoption 10 / 25
Maturity 15 / 25
Community 15 / 25

How are scores calculated?

Stars

198

Forks

22

Language

Python

License

MIT

Last pushed

Jul 07, 2025

Commits (30d)

0

Get this data via API

curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/MiniMax-AI/SynLogic"

Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.