base_model:
- BioMistral/BioMistral-7B
license: mit
pipeline_tag: text-generation
library_name: transformers
Mixture-Science-BioMistral-7B
🌐 Project Page | 💻 Code | 📄 Paper
We introduce RecursiveMAS, a multi-agent framework that scales agent collaboration through latent-space recursion. RecursiveMAS treats a multi-agent system as a unified recursive computation, where heterogeneous agents iteratively exchange, refine, and evolve their latent states across recursion rounds. In the Mixture-Style setting, the Science Specialist Agent focuses on science-oriented tasks and collaborates with other domain-specialized agents through RecursiveLink modules for final response generation.
Model Details
| Item | Description |
|---|---|
| Model | Mixture-Science-BioMistral-7B |
| Collaboration Style | Mixture-Style |
| Agent Role | Science Specialist Agent |
| Base Model | BioMistral-7B |
⚠️ Note: This checkpoint is a role-specific agent in RecursiveMAS, rather than a standalone model intended for plain-text generation.
Usage
To use this agent as part of the RecursiveMAS system, you can follow the instructions in the GitHub repository.
Programmatic Loading
You can load the multi-agent system (MAS) and access the specific science specialist agent using the provided high-level API:
from system_loader import load_mas_system
mas = load_mas_system(
style="mixture",
device="cuda",
trust_remote_code=True,
)
# Access the science agent model
science_agent = mas.agents["science"].model
CLI Inference
Alternatively, you can run inference for the Mixture-style collaboration pattern using the following command:
python run.py --style mixture --batch_size 16 --temperature 0.6 --top_p 0.95 --dataset math500 --seed 42 --trust_remote_code 1 --device cuda
Model Collections for RecursiveMAS
| Style | Model Collection |
|---|---|
| Sequential-Style | 🤗 HuggingFace |
| Mixture-Style | 🤗 HuggingFace |
| Distillation-Style | 🤗 HuggingFace |
| Deliberation-Style | 🤗 HuggingFace |
Experiment Results
Citation
@misc{recursivemas,
title={Recursive Multi-Agent Systems},
author={Xiyuan Yang and Jiaru Zou and Rui Pan and Ruizhong Qiu and Pan Lu and Shizhe Diao and Jindong Jiang and Hanghang Tong and Tong Zhang and Markus J. Buehler and Jingrui He and James Zou},
year={2026},
eprint={2604.25917},
archivePrefix={arXiv},
primaryClass={cs.AI},
url={https://arxiv.org/abs/2604.25917},
}