Hugging Face
Fine-tune SmolVLM on your data
Hugging Face's small SmolVLM model that reads images and text, which trains on one GPU. Supported by the training and serving software Tensorant runs. It hasn't run on Tensorant yet, so start with a trial.
- Status
- Supported
- Input
- Text and images
- Kind
- Dense
- Trains with
- LoRA, QLoRA
- Suggested sizes
- LoRA from1 GPU, about 9 GB
Sizes
Sizes and GPUs
Each size with the GPUs one machine needs to train and to serve it. The figures are the estimates Tensorant checks a model against; a trial measures the real use. See GPU memory.
| Model | Parameters | Train with LoRA | Train with QLoRA | Serve | Status |
|---|---|---|---|---|---|
| HuggingFaceTB/SmolVLM-Instruct | 2B | 1 GPUabout 9 GB | 1 GPUabout 6 GB | 1, 2, 4 or 8 GPUsabout 7 GB in all | Supported |
- Training needs the memory shown on each GPU: more GPUs on a machine train faster, but each holds the whole model.
- On a vision model, QLoRA trains with DPO only; SFT trains with LoRA.
- Serve: the numbers of GPUs of one machine the base model can be split across, and the memory it needs in all.
Training
How it trains
Train SmolVLM with LoRA or QLoRA, by supervised fine-tuning (SFT) or DPO on preference pairs.
- SFT trains with LoRA, on a version with images or of text alone. On text, the vision tower stays frozen and the served model still reads images.
- DPO trains with LoRA or QLoRA, on one GPU.
- RL trains text models only.
- SmolVLM trains on one GPU, on text or on images: its training doesn't split across a machine's GPUs.
Every training run, trials included, pays a training-token fee for the tokens it trains on, by the model's size. Images are on every organization.
Serving
How it serves
Publish a trained model under a name and call it through the OpenAI-compatible API, with the OpenAI SDK or any HTTP client. The name is what your application sends as model.
- A LoRA adapter you train serves on top of its base model, on the GPUs in the Serve column.
- The base model serves on its own too, in a test session or for a baseline. A deployment splits it across the GPUs of one machine: the Serve column lists the numbers each size allows.
- A served vision model reads images and text in the same chat request.
Steps
Fine-tune SmolVLM on Tensorant
Create a project
Create a project with
HuggingFaceTB/SmolVLM-Instructas its base model. The model check reads the model before anything starts.Import your data
Upload files, paste text or import a Hugging Face dataset as sources.
Review and freeze a version
Review the examples and approve the ones to train on, then freeze them into a version with training, validation and test splits.
Score the base model
Evaluate the base model on the version's test split. This baseline is what a training run is compared with, and a full run needs one.
Run a trial, then train
A trial trains a small sample to check the settings, the GPU memory and the speed. Then start the full run. LoRA on
HuggingFaceTB/SmolVLM-Instructneeds 1 GPU, about 9 GB.Compare and publish
Evaluate the trained model on the same test split and compare it with the baseline. Then publish it and call it through the OpenAI-compatible API.
Start with one dataset.
Create an account and import your data, or book a demo and we will show you the product.