HuggingFaceTB_SmolLM3-3B_0.py

# /// script
# requires-python = ">=3.12"
# dependencies = [
#     "numpy",
#     "einops",
#     "torch",
#     "transformers",
#     "diffusers",
#     "datasets",
#     "accelerate",
#     "timm",
# ]
# ///

try:
    # Use a pipeline as a high-level helper
    from transformers import pipeline
    from transformers import AutoTokenizer
    
    model_id = "HuggingFaceTB/SmolLM3-3B"
    
    tokenizer = AutoTokenizer.from_pretrained(model_id)
    
    pipe = pipeline("text-generation", model=model_id, tokenizer=tokenizer)
    
    messages = [
        {"role": "user", "content": "Give me a brief explanation of gravity in simple terms."},
    ]
    pipe(messages)
    
    messages = [
        {"role": "system", "content": "/no_think"},
        {"role": "user", "content": "Give me a brief explanation of gravity in simple terms."},
    ]
    pipe(messages)
    
    from transformers import AutoModelForCausalLM, AutoTokenizer
    
    model_name = "HuggingFaceTB/SmolLM3-3B"
    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 = "Give me a brief explanation of gravity in simple terms."
    messages_think = [
        {"role": "user", "content": prompt}
    ]
    
    text = tokenizer.apply_chat_template(
        messages_think,
        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))
    
    prompt = "Give me a brief explanation of gravity in simple terms."
    messages = [
        {"role": "system", "content": "/no_think"},
        {"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))
    
    tools = [
        {
            "name": "get_weather",
            "description": "Get the weather in a city",
            "parameters": {"type": "object", "properties": {"city": {"type": "string", "description": "The city to get the weather for"}}}}
    ]
    
    messages = [
        {
            "role": "user",
            "content": "Hello! How is the weather today in Copenhagen?"
        }
    ]
    
    inputs = tokenizer.apply_chat_template(
        messages,
        enable_thinking=False, # True works as well, your choice!
        xml_tools=tools,
        add_generation_prompt=True,
        tokenize=True,
        return_tensors="pt"
    ).to(model.device)
    
    outputs = model.generate(inputs)
    print(tokenizer.decode(outputs[0]))
    with open('HuggingFaceTB_SmolLM3-3B_0.txt', 'w') as f:
        f.write('Everything was good in HuggingFaceTB_SmolLM3-3B_0.txt')
except Exception as e:
    with open('HuggingFaceTB_SmolLM3-3B_0.txt', 'w') as f:
        import traceback
        traceback.print_exc(file=f)
finally:
    from huggingface_hub import upload_file
    upload_file(
        path_or_fileobj='HuggingFaceTB_SmolLM3-3B_0.txt',
        repo_id='model-metadata/custom_code_execution_files',
        path_in_repo='HuggingFaceTB_SmolLM3-3B_0.txt',
        repo_type='dataset',
    )