95 lines
3.5 KiB
Markdown
95 lines
3.5 KiB
Markdown
---
|
||
language:
|
||
- en
|
||
license: llama3
|
||
tags:
|
||
- Llama-3
|
||
- RL
|
||
- Atropos
|
||
- Tool Calling
|
||
- Nous Research
|
||
- instruct
|
||
- finetune
|
||
- reasoning
|
||
- function calling
|
||
- transformers
|
||
- reinforcement-learning
|
||
- json mode
|
||
- chatml
|
||
base_model: meta-llama/Meta-Llama-3.1-8B
|
||
library_name: transformers
|
||
|
||
---
|
||
|
||
# DeepHermes Tool Calling Specialist - Atropos RL
|
||
|
||
## Model Overview
|
||
|
||
The **DeepHermes Tool Calling Specialist - Atropos RL** model is an experimental artifact fine-tuned by Nous Research using our innovative open-source reinforcement learning framework—Atropos. This variant specifically improves the tool calling performance of the **DeepHermes 3 Llama-3.1 8B** model during its reasoning mode.
|
||
|
||
**Note**: This model is intended as an experimental artifact and is not designed for broad, general-purpose use.
|
||
|
||
## Atropos Open Source Framework
|
||
|
||
Atropos is Nous Research’s open-source Reinforcement Learning environment stack, designed to enhance various aspects of LLM functionalities through structured RL methodologies. We encourage contributions and exploration:
|
||
|
||
🔗 [Atropos GitHub Repository](https://github.com/NousResearch/Atropos)
|
||
|
||
## Benchmark Results
|
||
|
||
Evaluations on the Berkeley Function Calling benchmark demonstrate significant improvements in tool calling accuracy during reasoning mode, compared to its base model:
|
||
|
||
| Benchmark | Base Accuracy | Atropos RL Accuracy | Improvement |
|
||
| --------- | ------------- | ------------------- | ----------- |
|
||
| Parallel | 0.10 | 0.46 | **4.6x** |
|
||
| Simple | 0.21 | 0.5175 | **2.5x** |
|
||
|
||
These enhancements are due to RL fine-tuning specifically targeted at improving reasoning-based tool calling capabilities.
|
||
|
||
Eval set accuracy results:
|
||
|
||
|
||
|
||
## Key Features
|
||
|
||
- **Improved Tool Calling in Reasoning Mode**: Reinforcement learning significantly boosts tool usage during complex reasoning tasks.
|
||
- **Open-Source RL Framework**: Utilizes the fully open-source Atropos RL Environments.
|
||
- **Active Open Source Community**: Contributions welcomed on the Atropos GitHub.
|
||
- **Upcoming SOTA RL Trainer**: A state-of-the-art open-source reinforcement learning trainer by Nous Research is coming soon.
|
||
|
||
## Usage
|
||
|
||
This model supports multiple inference modes including:
|
||
|
||
- **Reasoning (Deep Thinking Mode)**
|
||
- **Standard Chat/Instruction Mode**
|
||
- **Structured JSON Outputs**
|
||
- **Function Calling**
|
||
|
||
Detailed documentation and example inference code are available:
|
||
|
||
*Note: You must first place DeepHermes' reasoning system prompt, and then append your function calling system prompt after for it to do reasoning and tool calling simultaneously.*
|
||
|
||
🔗 [Hermes Function Calling GitHub](https://github.com/NousResearch/Hermes-Function-Calling)
|
||
|
||
## How to Cite
|
||
|
||
```bibtex
|
||
@misc{
|
||
title={DeepHermes Tool Calling Specialist - Atropos RL},
|
||
author={Teknium and Dakota Mahan and Roger Jin and Chen Guang and Jai Suphavadeeprasit and Jeffrey Quesnelle},
|
||
year={2025},
|
||
url={https://huggingface.co/NousResearch/DeepHermes-Tool-Calling-Specialist-Atropos-RL}
|
||
}
|
||
```
|
||
|
||
## Community and Support
|
||
|
||
For questions, issues, or findings, please open issues or discussions in the respective GitHub repositories:
|
||
|
||
- [Atropos Framework Issues](https://github.com/NousResearch/Atropos/issues)
|
||
- [DeepHermes Models Issues](https://github.com/NousResearch/Hermes-Function-Calling/issues)
|
||
|
||
Nous Research encourages active community engagement and open-source contributions to continuously improve model performance and capabilities.
|
||
|