SmolLM3-3B-GGUF

SmolLM3-GGUF#

Original model: https://huggingface.co/HuggingFaceTB/SmolLM3-3B

Important. To enable thinking, you need to specify --jinja

Example usage with llama.cpp:

llama-cli -hf ggml-org/SmolLM3-3B-GGUF --jinja

Image: image/png

Table of Contents#

1. Model Summary
2. Evaluation
3. Training
4. Limitations
5. License

Model Summary#

SmolLM3 is a 3B parameter language model designed to push the boundaries of small models. It supports 6 languages, advanced reasoning and long context. SmolLM3 is a fully open model that offers strong performance at the 3B–4B scale.

Image: image/png

The model is a decoder-only transformer using GQA and NoRope, it was pretrained on 11.2T tokens with a staged curriculum of web, code, math and reasoning data. Post-training included midtraining on 140B reasoning tokens followed by supervised fine-tuning and alignment via Anchored Preference Optimization (APO).

Key features#

• Instruct model optimized for hybrid reasoning
• Fully open model: open weights + full training details including public data mixture and training configs
• Long context: Trained on 64k context and suppots up to 128k tokens using YARN extrapolation
• Multilingual: 6 natively supported (English, French, Spanish, German, Italian, and Portuguese)

For more details refer to our blog post: TODO

How to use#

The modeling code for SmolLM3 is available in transformers v4.53.0, so make sure to upgrade your transformers version. You can also load the model with the latest vllm which uses transformers as a backend.

pip install -U transformers

from transformers import AutoModelForCausalLM, AutoTokenizer

checkpoint = "HuggingFaceTB/SmolLM3-3B"
device = "cuda" # for GPU usage or "cpu" for CPU usage
tokenizer = AutoTokenizer.from_pretrained(checkpoint)
# for multiple GPUs install accelerate and do `model = AutoModelForCausalLM.from_pretrained(checkpoint, device_map="auto")`
model = AutoModelForCausalLM.from_pretrained(checkpoint).to(device)
inputs = tokenizer.encode("Gravity is", return_tensors="pt").to(device)
outputs = model.generate(inputs)
print(tokenizer.decode(outputs[0]))

For local inference, you can use llama.cpp, ONNX, MLX and MLC. You can find quantized checkpoints in this collection [TODO].

Evaluation#

In this section, we report the evaluation results of SmolLM3 base model. All evaluations are zero-shot unless stated otherwise, and we use lighteval to run them. For Ruler 64k evaluation, we apply YaRN to the Qwen models with 32k context to extrapolate the context length.

We highlight the best score in bold and underline the second-best score.

Base Pre-Trained Model#

English benchmarks#

Note: All evaluations are zero-shot unless stated otherwise.

Multilingual benchmarks#

The model has also been trained on Arabic (standard), Chinese and Russian data, but has seen fewer tokens in these languages compared to the 6 above. We report the performance on these langages for information.

Instruction Model#

No Extended Thinking#

Evaluation results of non reasoning models and reasoning models in no thinking mode. We highlight the best and second-best scores in bold.

Extended Thinking#

Evaluation results in reasoning mode for SmolLM3 and Qwen3 models:

Training#

Model#

• Architecture: Transformer decoder
• Pretraining tokens: 11T
• Precision: bfloat16

Software & hardware#

• GPUs: 384 H100
• Training Framework: nanotron
• Data processing framework: datatrove
• Evaluation framework: lighteval
• Post-training Framework: TRL

Open resources#

Here is an infographic with all the training details [TODO].

• The datasets used for pretraining can be found in this collection and those used in mid-training and pos-training can be found here [TODO]
• The training and evaluation configs and code can be found in the huggingface/smollm repository.

Limitations#

SmolLM3 can produce text on a variety of topics, but the generated content may not always be factually accurate, logically consistent, or free from biases present in the training data. These models should be used as assistive tools rather than definitive sources of information. Users should always verify important information and critically evaluate any generated content.

License#

Apache 2.0