PRIME-RL/PRIME
Scalable RL solution for advanced reasoning of language models
This project helps AI researchers and machine learning engineers significantly improve the reasoning abilities of large language models (LLMs) for complex tasks like coding and math. It takes an existing LLM and task-specific data, then applies a reinforcement learning approach to output a more capable LLM that can generate better, more accurate solutions. It's designed for those who develop and fine-tune advanced AI models.
1,813 stars. No commits in the last 6 months.
Use this if you need to enhance the reasoning and problem-solving capabilities of your large language models beyond what standard training methods can achieve, especially for tasks requiring step-by-step logical thought.
Not ideal if you are a general user looking for an off-the-shelf application or if you lack experience with advanced machine learning concepts and model training workflows.
Stars
1,813
Forks
104
Language
Python
License
Apache-2.0
Category
Last pushed
Mar 18, 2025
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/PRIME-RL/PRIME"
Open to everyone — 100 requests/day, no key needed. Get a free key for 1,000/day.
Higher-rated alternatives
open-thought/reasoning-gym
[NeurIPS 2025 Spotlight] Reasoning Environments for Reinforcement Learning with Verifiable Rewards
Hmbown/Hegelion
Dialectical reasoning architecture for LLMs (Thesis → Antithesis → Synthesis)
LLM360/Reasoning360
A repo for open research on building large reasoning models
TsinghuaC3I/Awesome-RL-for-LRMs
A Survey of Reinforcement Learning for Large Reasoning Models
bowang-lab/BioReason
BioReason: Incentivizing Multimodal Biological Reasoning within a DNA-LLM Model | NeurIPS '25