Instructions to use NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos with Transformers:

# Use a pipeline as a high-level helper
from transformers import pipeline

pipe = pipeline("text-generation", model="NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos")
model = AutoModelForCausalLM.from_pretrained("NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos")
messages = [
    {"role": "user", "content": "Who are you?"},
]
inputs = tokenizer.apply_chat_template(
	messages,
	add_generation_prompt=True,
	tokenize=True,
	return_dict=True,
	return_tensors="pt",
).to(model.device)

outputs = model.generate(**inputs, max_new_tokens=40)
print(tokenizer.decode(outputs[0][inputs["input_ids"].shape[-1]:]))

Inference
Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos

SGLang

How to use NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos with SGLang:

Install from pip and serve model

# Install SGLang from pip:
pip install sglang
# Start the SGLang server:
python3 -m sglang.launch_server \
    --model-path "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos" \
    --host 0.0.0.0 \
    --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker images

docker run --gpus all \
    --shm-size 32g \
    -p 30000:30000 \
    -v ~/.cache/huggingface:/root/.cache/huggingface \
    --env "HF_TOKEN=<secret>" \
    --ipc=host \
    lmsysorg/sglang:latest \
    python3 -m sglang.launch_server \
        --model-path "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos" \
        --host 0.0.0.0 \
        --port 30000
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:30000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos with Docker Model Runner:
```
docker model run hf.co/NousResearch/DeepHermes-Financial-Fundamentals-Prediction-Specialist-Atropos
```
Browse Quantizations to use this model in llama.cpp, Ollama, LM Studio, or any compatible app.

DeepHermes Financial Fundamentals Prediction Specialist - Atropos RL

Model Overview

The DeepHermes Financial Fundamentals Prediction Specialist - Atropos RL is an experimental model artifact, fine-tuned by Nous Research using our new open source LLM RL Gym, Atropos. This model specifically aims to enhance the accuracy of financial fundamentals predictions through reasoning-intensive reinforcement learning techniques.

Note: This model is experimental and is not intended as a general-purpose state-of-the-art solution.

Atropos Open Source Framework

Atropos is Nous Research's open-source reinforcement learning environment stack, engineered to optimize diverse LLM capabilities through structured RL methodologies. Contributions and active engagement from the community are highly encouraged:

🔗 Atropos GitHub Repository

Evaluation Results

Training and evaluation focused on improving financial fundamentals direction prediction accuracy:

Evaluation Metric	Final Accuracy
Direction Prediction Accuracy (train)	~20% -> ~50% Accuracy

Key Features

Financial Fundamentals Prediction: Enhanced capability to predict market direction using RL during intensive reasoning tasks.
Built with Atropos RL Environments: Built with the open-source Atropos reinforcement learning stack.
Reasoning-Enhanced Predictions: Specifically optimized for scenarios involving deep analytical reasoning in financial contexts.

Usage

The model is optimized for reasoning-intensive financial analysis tasks and supports:

Deep Reasoning Mode given a company's context (previous quarter's financials data) to predict the next future quarter's fundamental metric direction)

Community and Support

We welcome contributions, suggestions, and issues through our community channels:

Atropos Issues

How to Cite

@misc{
      title={DeepHermes Financial Fundamentals Prediction 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-Financial-Fundamentals-Prediction-Specialist-Atropos-RL}
}