Add pipeline_tag, library_name and sample usage
#1
by nielsr HF Staff - opened
README.md
CHANGED
|
@@ -2,6 +2,8 @@
|
|
| 2 |
base_model:
|
| 3 |
- meta-llama/Llama-3.2-1B-Instruct
|
| 4 |
license: mit
|
|
|
|
|
|
|
| 5 |
---
|
| 6 |
|
| 7 |
# Sequential-Light-Critic-Llama3.2-1B
|
|
@@ -21,7 +23,26 @@ RecursiveMAS treats a multi-agent system as a unified recursive computation, whe
|
|
| 21 |
| Base Model | Llama3.2-1B |
|
| 22 |
|
| 23 |
⚠️ **Note:** This checkpoint is a **role-specific agent** in [**RecursiveMAS**](https://arxiv.org/abs/2604.25917), rather than a standalone model intended for plain-text generation.
|
| 24 |
-
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
|
| 25 |
|
| 26 |
## Model Collections for RecursiveMAS
|
| 27 |
|
|
@@ -43,7 +64,7 @@ For detailed usage instructions, please refer to our [GitHub repository](https:/
|
|
| 43 |
```bibtex
|
| 44 |
@misc{recursivemas,
|
| 45 |
title={Recursive Multi-Agent Systems},
|
| 46 |
-
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},
|
| 47 |
year={2026},
|
| 48 |
eprint={2604.25917},
|
| 49 |
archivePrefix={arXiv},
|
|
|
|
| 2 |
base_model:
|
| 3 |
- meta-llama/Llama-3.2-1B-Instruct
|
| 4 |
license: mit
|
| 5 |
+
pipeline_tag: text-generation
|
| 6 |
+
library_name: transformers
|
| 7 |
---
|
| 8 |
|
| 9 |
# Sequential-Light-Critic-Llama3.2-1B
|
|
|
|
| 23 |
| Base Model | Llama3.2-1B |
|
| 24 |
|
| 25 |
⚠️ **Note:** This checkpoint is a **role-specific agent** in [**RecursiveMAS**](https://arxiv.org/abs/2604.25917), rather than a standalone model intended for plain-text generation.
|
| 26 |
+
|
| 27 |
+
## Sample Usage
|
| 28 |
+
|
| 29 |
+
You can load the entire MAS pipeline using the code provided in the [GitHub repository](https://github.com/RecursiveMAS/RecursiveMAS):
|
| 30 |
+
|
| 31 |
+
```python
|
| 32 |
+
from system_loader import load_mas_system
|
| 33 |
+
|
| 34 |
+
mas = load_mas_system(
|
| 35 |
+
style="sequential_light",
|
| 36 |
+
device="cuda",
|
| 37 |
+
trust_remote_code=True,
|
| 38 |
+
)
|
| 39 |
+
|
| 40 |
+
planner = mas.agents["planner"].model
|
| 41 |
+
critic = mas.agents["critic"].model
|
| 42 |
+
solver = mas.agents["solver"].model
|
| 43 |
+
```
|
| 44 |
+
|
| 45 |
+
For detailed usage instructions, please refer to the [GitHub repository](https://github.com/RecursiveMAS/RecursiveMAS).
|
| 46 |
|
| 47 |
## Model Collections for RecursiveMAS
|
| 48 |
|
|
|
|
| 64 |
```bibtex
|
| 65 |
@misc{recursivemas,
|
| 66 |
title={Recursive Multi-Agent Systems},
|
| 67 |
+
author={Xiyuan Yang and Jiaru Zou and Rui Pan and Ruizhong Qiu and Pan Lu front and Shizhe Diao and Jindong Jiang and Hanghang Tong and Tong Zhang and Markus J. Buehler and Jingrui He and James Zou},
|
| 68 |
year={2026},
|
| 69 |
eprint={2604.25917},
|
| 70 |
archivePrefix={arXiv},
|