Instructions to use CastIronMind/stentor_python_30m with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use CastIronMind/stentor_python_30m with Transformers:

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

pipe = pipeline("text-generation", model="CastIronMind/stentor_python_30m")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("CastIronMind/stentor_python_30m")
model = AutoModelForCausalLM.from_pretrained("CastIronMind/stentor_python_30m")
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

vLLM

How to use CastIronMind/stentor_python_30m with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "CastIronMind/stentor_python_30m"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "CastIronMind/stentor_python_30m",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/CastIronMind/stentor_python_30m

SGLang

How to use CastIronMind/stentor_python_30m 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 "CastIronMind/stentor_python_30m" \
    --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": "CastIronMind/stentor_python_30m",
		"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 "CastIronMind/stentor_python_30m" \
        --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": "CastIronMind/stentor_python_30m",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use CastIronMind/stentor_python_30m with Docker Model Runner:
```
docker model run hf.co/CastIronMind/stentor_python_30m
```

Stanislav commited on Mar 3

Commit

e858b1c

verified ·

1 Parent(s): feed67b

Update README.md

Browse files

Files changed (1) hide show

README.md +2 -22

README.md CHANGED Viewed

@@ -35,8 +35,8 @@ All examples follow a consistent format with "### Task:" instruction and "### So
 The fine-tuning process involved multiple stages:
-1. Base model: Stentor-30M pre-trained checkpoint
-2. Initial fine-tuning on 50k examples (checkpoint-1000 selected as best)
 3. Multiple correction rounds with progressively lower learning rates
 4. Final detoxification training with learning rate 3e-7 to remove undesirable patterns
@@ -100,30 +100,10 @@ outputs = model.generate(**inputs, max_new_tokens=100, temperature=0.2)
 response = tokenizer.decode(outputs[0], skip_special_tokens=True)
 ```
-**Hardware Requirements**
-- **Inference:** CPU only (no GPU required)
-- **RAM:** < 100 MB for inference
-- **Storage:** 60 MB (FP16), 30 MB (INT8 quantized)
 **Ethical Considerations**
 This model is intended for educational and development assistance purposes. Users should verify all generated code before deployment, particularly for security-sensitive applications. The model may occasionally produce incorrect or inefficient code and should not be relied upon as the sole source of truth for programming tasks.
-**Citation**
-If you use this model in your work, please cite:
-```
-@misc{stentor-python-30m-2026,
-  author = {Fine-tuning Experiment},
-  title = {Stentor Python 30M: A Compact Model for Python Code Generation},
-  year = {2026},
-  publisher = {Hugging Face},
-  url = {https://huggingface.co/username/stentor-python-30m}
-}
-```
 **Contact**
 For questions or feedback about this model, please open an issue on the Hugging Face repository.

 The fine-tuning process involved multiple stages:
+1. Base model: Stentor-30M
+2. Initial fine-tuning on 50k examples
 3. Multiple correction rounds with progressively lower learning rates
 4. Final detoxification training with learning rate 3e-7 to remove undesirable patterns
 response = tokenizer.decode(outputs[0], skip_special_tokens=True)
 ```
 **Ethical Considerations**
 This model is intended for educational and development assistance purposes. Users should verify all generated code before deployment, particularly for security-sensitive applications. The model may occasionally produce incorrect or inefficient code and should not be relied upon as the sole source of truth for programming tasks.
 **Contact**
 For questions or feedback about this model, please open an issue on the Hugging Face repository.