PPPP-kaqiu/Awesome-Parallel-Reasoning
Awesome-Parallel-Reasoning: Unlocking the reasoning potential of LLMs. Papers, Code, Resources & Survey.
This resource collects information on an emerging technique called 'parallel reasoning' for Large Language Models (LLMs). It explains how generating and evaluating multiple reasoning paths simultaneously, rather than a single one, can lead to more robust, higher-quality outputs and significantly faster response times. This is for researchers, engineers, and product managers working to improve the performance and user experience of LLM-powered applications.
Use this if you are exploring ways to make your LLM applications faster, more reliable, and capable of complex problem-solving by leveraging simultaneous processing.
Not ideal if you are looking for an off-the-shelf software solution or a simple API to integrate, as this project focuses on research and conceptual frameworks.
Stars
49
Forks
4
Language
HTML
License
MIT
Category
Last pushed
Mar 08, 2026
Commits (30d)
0
Get this data via API
curl "https://pt-edge.onrender.com/api/v1/quality/llm-tools/PPPP-kaqiu/Awesome-Parallel-Reasoning"
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