SmolVLM2-2.2B is a lightweight multimodal model designed to analyze video content. The model processes videos, images, and text inputs to generate text outputs - whether answering questions about media files, comparing visual content, or transcribing text from images. Despite its compact size, requiring only 5.2GB of GPU RAM for video inference, it delivers robust performance on complex multimodal tasks. This efficiency makes it particularly well-suited for on-device applications where computational resources may be limited.
Model Summary
Developed by: Hugging Face 🤗
Model type: Multi-modal model (image/multi-image/video/text)
Language(s) (NLP): English
License: Apache 2.0
Architecture: Based on Idefics3 (see technical summary)
SmolVLM2 can be used for inference on multimodal (video / image / text) tasks where the input consists of text queries along with video or one or more images. Text and media files can be interleaved arbitrarily, enabling tasks like captioning, visual question answering, and storytelling based on visual content. The model does not support image or video generation.
We evaluated the performance of the SmolVLM2 family on the following scientific benchmarks:
Size
Video-MME
MLVU
MVBench
2.2B
52.1
55.2
46.27
500M
42.2
47.3
39.73
256M
33.7
40.6
32.7
How to get started
You can use transformers to load, infer and fine-tune SmolVLM. Make sure you have num2words, flash-attn and latest transformers installed.
You can load the model as follows.
To use SmolVLM2 for video inference, make sure you have decord installed.
python
1messages =[2{3"role":"user",4"content":[5{"type":"video","path":"path_to_video.mp4"},6{"type":"text","text":"Describe this video in detail"}7]8},9]1011inputs = processor.apply_chat_template(12 messages,13 add_generation_prompt=True,14 tokenize=True,15 return_dict=True,16 return_tensors="pt",17).to(model.device, dtype=torch.bfloat16)1819generated_ids = model.generate(**inputs, do_sample=False, max_new_tokens=64)20generated_texts = processor.batch_decode(21 generated_ids,22 skip_special_tokens=True,23)2425print(generated_texts[0])
Multi-image Interleaved Inference
You can interleave multiple media with text using chat templates.
python
1import torch
234messages =[5{6"role":"user",7"content":[8{"type":"text","text":"What is the similarity between these two images?"},9{"type":"image","url":"https://huggingface.co/datasets/huggingface/documentation-images/resolve/main/bee.jpg"},10{"type":"image","url":"https://huggingface.co/datasets/huggingface/documentation-images/resolve/0052a70beed5bf71b92610a43a52df6d286cd5f3/diffusers/rabbit.jpg"},11]12},13]1415inputs = processor.apply_chat_template(16 messages,17 add_generation_prompt=True,18 tokenize=True,19 return_dict=True,20 return_tensors="pt",21).to(model.device, dtype=torch.bfloat16)2223generated_ids = model.generate(**inputs, do_sample=False, max_new_tokens=64)24generated_texts = processor.batch_decode(25 generated_ids,26 skip_special_tokens=True,27)28print(generated_texts[0])
Model optimizations
Misuse and Out-of-scope Use
SmolVLM is not intended for high-stakes scenarios or critical decision-making processes that affect an individual's well-being or livelihood. The model may produce content that appears factual but may not be accurate. Misuse includes, but is not limited to:
Prohibited Uses:
Evaluating or scoring individuals (e.g., in employment, education, credit)
We release the SmolVLM2 checkpoints under the Apache 2.0 license.
Citation information
You can cite us in the following way:
bibtex
1@article{marafioti2025smolvlm,
2 title={SmolVLM: Redefining small and efficient multimodal models},
3 author={Andrés Marafioti and Orr Zohar and Miquel Farré and Merve Noyan and Elie Bakouch and Pedro Cuenca and Cyril Zakka and Loubna Ben Allal and Anton Lozhkov and Nouamane Tazi and Vaibhav Srivastav and Joshua Lochner and Hugo Larcher and Mathieu Morlon and Lewis Tunstall and Leandro von Werra and Thomas Wolf},
4 journal={arXiv preprint arXiv:2504.05299},
5 year={2025}
6}
