Instructions to use lewtun/SmolLM2-360M-OpenMathReasoning with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use lewtun/SmolLM2-360M-OpenMathReasoning with Transformers:

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

pipe = pipeline("text-generation", model="lewtun/SmolLM2-360M-OpenMathReasoning")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("lewtun/SmolLM2-360M-OpenMathReasoning")
model = AutoModelForCausalLM.from_pretrained("lewtun/SmolLM2-360M-OpenMathReasoning")
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]:]))

Notebooks
Google Colab
Kaggle
Local Apps Settings

vLLM

How to use lewtun/SmolLM2-360M-OpenMathReasoning with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "lewtun/SmolLM2-360M-OpenMathReasoning"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "lewtun/SmolLM2-360M-OpenMathReasoning",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/lewtun/SmolLM2-360M-OpenMathReasoning

SGLang

How to use lewtun/SmolLM2-360M-OpenMathReasoning 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 "lewtun/SmolLM2-360M-OpenMathReasoning" \
    --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": "lewtun/SmolLM2-360M-OpenMathReasoning",
		"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 "lewtun/SmolLM2-360M-OpenMathReasoning" \
        --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": "lewtun/SmolLM2-360M-OpenMathReasoning",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use lewtun/SmolLM2-360M-OpenMathReasoning with Docker Model Runner:
```
docker model run hf.co/lewtun/SmolLM2-360M-OpenMathReasoning
```

SmolLM2-360M-OpenMathReasoning

File size: 1,947 Bytes

4045ceb
8bc11bd
 
 
4045ceb
8bc11bd
 
 
 
 
 
4045ceb
 
8bc11bd
4045ceb
8bc11bd
 
4045ceb
8bc11bd
4045ceb
 
8bc11bd
4045ceb
8bc11bd
 
 
 
4045ceb
 
8bc11bd
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
eab4232

---
base_model: HuggingFaceTB/SmolLM2-360M-Instruct
library_name: transformers
model_name: SmolLM2-360M-OpenMathReasoning
tags:
- generated_from_trainer
- sft
- hf_jobs
- trackio:https://lewtun-mlintern-omr360m1.hf.space?project=huggingface&runs=smollm2-360m-openmath-cot-lr2e5-bs16&sidebar=collapsed
- trl
licence: license
---

# Model Card for SmolLM2-360M-OpenMathReasoning

This model is a fine-tuned version of [HuggingFaceTB/SmolLM2-360M-Instruct](https://huggingface.co/HuggingFaceTB/SmolLM2-360M-Instruct).
It has been trained using [TRL](https://github.com/huggingface/trl).

## Quick start

```python
from transformers import pipeline

question = "If you had a time machine, but could only go to the past or the future once and never return, which would you choose and why?"
generator = pipeline("text-generation", model="lewtun/SmolLM2-360M-OpenMathReasoning", device="cuda")
output = generator([{"role": "user", "content": question}], max_new_tokens=128, return_full_text=False)[0]
print(output["generated_text"])
```

## Training procedure

 
[<img src="https://raw.githubusercontent.com/gradio-app/trackio/refs/heads/main/trackio/assets/badge.png" alt="Visualize in Trackio" title="Visualize in Trackio" width="150" height="24"/>](https://lewtun-mlintern-omr360m1.hf.space?project=huggingface&runs=smollm2-360m-openmath-cot-lr2e5-bs16&sidebar=collapsed)


This model was trained with SFT.

### Framework versions

- TRL: 1.4.0
- Transformers: 5.8.0
- Pytorch: 2.11.0
- Datasets: 4.8.5
- Tokenizers: 0.22.2

## Citations



Cite TRL as:
    
```bibtex
@software{vonwerra2020trl,
  title   = {{TRL: Transformers Reinforcement Learning}},
  author  = {von Werra, Leandro and Belkada, Younes and Tunstall, Lewis and Beeching, Edward and Thrush, Tristan and Lambert, Nathan and Huang, Shengyi and Rasul, Kashif and Gallouédec, Quentin},
  license = {Apache-2.0},
  url     = {https://github.com/huggingface/trl},
  year    = {2020}
}
```