26 dependents
| Package | Description | Downloads/month |
|---|---|---|
| Our library for RL environments + evals | 2.3M | |
| This is an open-source version of the representation engineering framework for s... | 167K | |
| mellea is a library for writing generative programs | 33K | |
| One-for-All Multimodal Evaluation Toolkit Across Text, Image, Video, and Audio T... | 16K | |
| OpenCompass VLM Evaluation Kit for Eval-Scope | 15K | |
| OpenJudge: A Unified Framework for Holistic Evaluation and Quality Rewards | 11K | |
| A Gym for Agentic LLMs | 6K | |
| 6K | ||
| Trinity-RFT is a general-purpose, flexible and scalable framework designed for r... | 3K | |
| Modern Data Centric AI system for Large Language Models | 3K | |
| Atropos: An Environment and Rollout handler for LLM RL | 2K | |
| Cosmos-RL is a flexible and scalable Reinforcement Learning framework specialize... | 2K | |
| 2K | ||
| Standardizing environment infrastructure with Strands Agents — step, observe, re... | 1K | |
| Easy Data Preparation with latest LLMs-based Operators and Pipelines. | 1K | |
| 🌾 OAT: A research-friendly framework for LLM online alignment, including reinfor... | 788 | |
| 532 | ||
| 405 | ||
| siiRL: Shanghai Innovation Institute RL Framework for Advanced LLMs and Multi-Ag... | 348 | |
| [ICML26 Spotlight] UniPercept: Towards Unified Perceptual-Level Image Understand... | 334 | |
| The most modern LLM evaluation toolkit | 219 | |
| MMATH - packaged by NVIDIA | 212 | |
| Export Camel toolkits as MCP servers | 193 | |
| Multi-Turn Multi-Agent RL with LLMs for the GPU poor and middle class. Supports... | 168 | |
| ModelSelect: A Unified Framework for Holistic Evaluation and Quality Reward | 161 | |
| nanoRLHF: from-scratch journey into how LLMs and RLHF really work. | 123 |
