Instructions to use alphaedge-ai/SmolLM3-3B-ita-32768 with libraries, inference providers, notebooks, and local apps. Follow these links to get started.

Libraries

How to use alphaedge-ai/SmolLM3-3B-ita-32768 with Transformers:

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

pipe = pipeline("text-generation", model="alphaedge-ai/SmolLM3-3B-ita-32768")
messages = [
    {"role": "user", "content": "Who are you?"},
]
pipe(messages)

# Load model directly
from transformers import AutoTokenizer, AutoModelForCausalLM

tokenizer = AutoTokenizer.from_pretrained("alphaedge-ai/SmolLM3-3B-ita-32768")
model = AutoModelForCausalLM.from_pretrained("alphaedge-ai/SmolLM3-3B-ita-32768")
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 alphaedge-ai/SmolLM3-3B-ita-32768 with vLLM:

Install from pip and serve model

# Install vLLM from pip:
pip install vllm
# Start the vLLM server:
vllm serve "alphaedge-ai/SmolLM3-3B-ita-32768"
# Call the server using curl (OpenAI-compatible API):
curl -X POST "http://localhost:8000/v1/chat/completions" \
	-H "Content-Type: application/json" \
	--data '{
		"model": "alphaedge-ai/SmolLM3-3B-ita-32768",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Use Docker

docker model run hf.co/alphaedge-ai/SmolLM3-3B-ita-32768

SGLang

How to use alphaedge-ai/SmolLM3-3B-ita-32768 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 "alphaedge-ai/SmolLM3-3B-ita-32768" \
    --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": "alphaedge-ai/SmolLM3-3B-ita-32768",
		"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 "alphaedge-ai/SmolLM3-3B-ita-32768" \
        --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": "alphaedge-ai/SmolLM3-3B-ita-32768",
		"messages": [
			{
				"role": "user",
				"content": "What is the capital of France?"
			}
		]
	}'

Docker Model Runner
How to use alphaedge-ai/SmolLM3-3B-ita-32768 with Docker Model Runner:
```
docker model run hf.co/alphaedge-ai/SmolLM3-3B-ita-32768
```

SmolLM3-3B-ita-32768

This model is a 6.36% smaller version of HuggingFaceTB/SmolLM3-3B optimized for Italian language via vocabulary size reduction using the trimming method.
This trimmed model should perform similarly to the original model with only 32,768 tokens and a much smaller memory footprint. However, it may not perform well for other languages as tokens not commonly used in the selected languages were removed from the vocabulary.

Model Statistics

Metric	Original	Trimmed	Reduction
Vocabulary size	128,256 tokens	32,768 tokens	74.45%
Model size	3,075,098,624 params	2,879,539,200 params	6.36%

Mining Dataset Statistics

Number of texts used for mining: 200,000 texts
Dataset: lbourdois/fineweb-2-trimming

Usage

from transformers import AutoModelForCausalLM, AutoTokenizer

model_name = "alphaedge-ai/SmolLM3-3B-ita-32768"
device = "cuda"  # for GPU usage or "cpu" for CPU usage

# load the tokenizer and the model
tokenizer = AutoTokenizer.from_pretrained(model_name)
model = AutoModelForCausalLM.from_pretrained(
    model_name,
).to(device)

# prepare the model input
prompt = "Your prompt in Italian."
messages = [
    {"role": "user", "content": prompt}
]
text = tokenizer.apply_chat_template(
    messages,
    tokenize=False,
    add_generation_prompt=True,
)
model_inputs = tokenizer([text], return_tensors="pt").to(model.device)

# Generate the output
generated_ids = model.generate(**model_inputs, max_new_tokens=32768)

# Get and decode the output
output_ids = generated_ids[0][len(model_inputs.input_ids[0]):]
print(tokenizer.decode(output_ids, skip_special_tokens=True))

To enable/disable thinking mode, use the /think or /no_think flag in the system prompt:

messages = [
    {"role": "system", "content": "/no_think"},
    {"role": "user", "content": prompt}
]

Citations

SmolLM3-3B

@misc{bakouch2025smollm3,
      title={SmolLM3: smol, multilingual, long-context reasoner},
      author={akouch, Elie and Ben Allal, Loubna and Lozhkov, Anton and Tazi, Nouamane
               and Tunstall, Lewis and Patiño, Carlos Miguel and Beeching, Edward
               and Roucher, Aymeric and others},
      year={2025},
      howpublished={https://huggingface.co/blog/smollm3}
}

Trimming blog post

@misc{hf_blogpost_trimming,
      title={Introduction to Trimming}, 
      author={Loïck BOURDOIS and Tom AARSEN and Bram VANROY and Christopher AKIKI and Woojun JUNG and Manuel ROMERO and Prithiv SAKTHI},
      year={2026},
      url={https://huggingface.co/blog/lbourdois/introduction-to-trimming}, 
}