What are the best open-source alternatives to PROF?

Q: Is kwai-klear/klearreasoner a good alternative to PROF?

[December 5, 2025] 🔍 We propose entropy ratio clipping (ERC) to impose a global constraint on the output distribution of the policy model. Experiments demonstrate that ERC can significantly improve the stability of off-policy training. 📄 The paper is available on arXiv.

Question 1

Accepted Answer

19 open-source projects similar to chenluye99/prof, ranked by shared features. Top picks: aiframeresearch/spo, amap-ml/tree-grpo, cjreinforce/pure, cmu-aire/mrt, dongguanting/tool-star, gen-verse/reasonflux, kwai-klear/klearreasoner, langfengq/verl-agent, mcgill-nlp/vineppo, multimodal-art-projection/treepo.

Question 2

Is aiframeresearch/spo a good alternative to PROF?

Accepted Answer

🚀 Segment Policy Optimization: Effective Segment-Level Credit Assignment in RL for Large Language Models 🌟

Question 3

Is amap-ml/tree-grpo a good alternative to PROF?

Accepted Answer

Tree Search for LLM Agent Reinforcement Learning

Question 4

Is cjreinforce/pure a good alternative to PROF?

Accepted Answer

[2025/10/23] 🔥🔥Our paper is accepted by NeurIPS 2025.🔥🔥 - [2025/04/22] Released our Paper on arXiv. See here - [2025/03/24] We re-implement our algorithm based on verl. ✨✨ Key features: (1) add ~50 additional metrics to comprehensively monitor the training process and stability, (2) add a…

Question 5

Is cmu-aire/mrt a good alternative to PROF?

Accepted Answer

This repository contains the code for our paper titled "Optimizing Test-Time Compute via Meta Reinforcement Finetuning." In this work, we introduce a novel approach to optimizing test-time compute through meta reinforcement learning, aiming to balance the efficiency and discovery capabilities of…

Question 6

Is dongguanting/tool-star a good alternative to PROF?

Accepted Answer

🔧✨Tool-Star: Empowering Multi-Tool Collaborative Web Agent via Reinforcement Learning

Question 7

Is gen-verse/reasonflux a good alternative to PROF?

Accepted Answer

Princeton University \& PKU \& UIUC \& University of Chicago \& ByteDance Seed

Question 8

Is kwai-klear/klearreasoner a good alternative to PROF?

Accepted Answer

[December 5, 2025] 🔍 We propose entropy ratio clipping​ (ERC) to impose a global constraint on the output distribution of the policy model. Experiments demonstrate that ERC can significantly improve the stability of off-policy training. 📄 The paper is available on arXiv.

Question 9

Is langfengq/verl-agent a good alternative to PROF?

Accepted Answer

langfengq/verl-agent is an open-source alternative to PROF.

Question 10

Is mcgill-nlp/vineppo a good alternative to PROF?

Accepted Answer

Paper - Abstract - Updates - Quick Start - Installation - Download the datasets - Create Experiment Script - Single GPU Training (Only for Rho models) - Running the experiments - Code Structure - Initial SFT Checkpoints - Acknowledgement - Citation

Question 11

Is multimodal-art-projection/treepo a good alternative to PROF?

Accepted Answer

This is the official implementation of TreePO algorithm.

Open-source alternatives to PROF

AIFrameResearch/SPO

AMAP-ML/Tree-GRPO

CJReinforce/PURE

CMU-AIRe/MRT

dongguanting/Tool-Star

Open-source alternatives to PROF

AIFrameResearch/SPO

AMAP-ML/Tree-GRPO

CJReinforce/PURE

CMU-AIRe/MRT

dongguanting/Tool-Star

Gen-Verse/ReasonFlux

Kwai-Klear/KlearReasoner

langfengQ/verl-agent

McGill-NLP/VinePPO

multimodal-art-projection/TreePO

openai/prm800k

PRIME-RL/ImplicitPRM

PRIME-RL/PRIME

RUC-NLPIR/ARPO

RyanLiu112/AttnRL

THUDM/TreeRL

WangHanLinHenry/SPA-RL-Agent

yangzhch6/TreeRPO

Zillwang/StepSearch