Discover the power of compact language models with HuggingFaceTB's SmolLM3-3B
Optimized for efficiency without sacrificing performance. Perfect for edge devices and constrained environments.
Capable of handling diverse tasks from text generation to code completion with impressive accuracy.
Built on open standards and available for research and commercial use under permissive licenses.
Base Model
The core language model with 3 billion parameters optimized for efficient inference.
Instruction-Tuned
Fine-tuned version optimized for following instructions and conversational AI applications.
from transformers import ( AutoModelForCausalLM, AutoTokenizer, ) import torch model_id = "HuggingFaceTB/SmolLM3-3B" model = AutoModelForCausalLM.from_pretrained( model_id, torch_dtype=torch.float16, device_map="auto" ) tokenizer = AutoTokenizer.from_pretrained(model_id) # Generate text input_text = "Explain quantum computing in simple terms" inputs = tokenizer(input_text, return_tensors="pt").to("cuda") outputs = model.generate(**inputs, max_new_tokens=100) print(tokenizer.decode(outputs[0], skip_special_tokens=True))
Deploy conversational agents with low latency requirements
Integrate coding suggestions into development environments
Generate articles, stories, and marketing copy efficiently
Enable NLP capabilities in academic and scientific applications
Join thousands of developers exploring the potential of compact language models