GLM-5.3 is live. Run Z.ai's latest model on Friendli Model APIs. Try it today ➜
HuggingFaceTB
Available on FriendliAI
Run this model inference on single tenant GPU with unmatched speed and reliability at scale.
Model Details
Model Provider
HuggingFaceTB
Model Tree
HuggingFaceTB/SmolLM2-1.7B-Instruct
Input Modalities
Output Modalities
Supported Functionality
SmolVLM can be used for inference on multimodal (image + text) tasks where the input comprises text queries along with one or more images. Text and images can be interleaved arbitrarily, enabling tasks like image captioning, visual question answering, and storytelling based on visual content. The model does not support image generation.
To fine-tune SmolVLM on a specific task, you can follow the fine-tuning tutorial.
SmolVLM leverages the lightweight SmolLM2 language model to provide a compact yet powerful multimodal experience. It introduces several changes compared to previous Idefics models:
More details about the training and architecture are available in our technical report.
You can use transformers to load, infer and fine-tune SmolVLM.
python
import torchfrom PIL import Imagefrom transformers import AutoProcessor, AutoModelForVision2Seqfrom transformers.image_utils import load_imageDEVICE = "cuda" if torch.cuda.is_available() else "cpu"# Load imagesimage1 = load_image("https://cdn.britannica.com/61/93061-050-99147DCE/Statue-of-Liberty-Island-New-York-Bay.jpg")image2 = load_image("https://huggingface.co/spaces/merve/chameleon-7b/resolve/main/bee.jpg")# Initialize processor and modelprocessor = AutoProcessor.from_pretrained("HuggingFaceTB/SmolVLM-Base")model = AutoModelForVision2Seq.from_pretrained( "HuggingFaceTB/SmolVLM-Base", torch_dtype=torch.bfloat16, _attn_implementation="flash_attention_2" if DEVICE == "cuda" else "eager",).to(DEVICE)# Create input messagesmessages = [ { "role": "user", "content": [ {"type": "image"}, {"type": "image"}, {"type": "text", "text": "Can you describe the two images?"} ] },]# Prepare inputsprompt = processor.apply_chat_template(messages, add_generation_prompt=True)inputs = processor(text=prompt, images=[image1, image2], return_tensors="pt")inputs = inputs.to(DEVICE)# Generate outputsgenerated_ids = model.generate(**inputs, max_new_tokens=500)generated_texts = processor.batch_decode( generated_ids, skip_special_tokens=True,)print(generated_texts[0])"""User:<image>Can you describe the two images?Assistant: I can describe the first one, but I can't describe the second one."""Precision: For better performance, load and run the model in half-precision (torch.float16 or torch.bfloat16) if your hardware supports it.
python
from transformers import AutoModelForVision2Seqimport torchmodel = AutoModelForVision2Seq.from_pretrained( "HuggingFaceTB/SmolVLM-Base", torch_dtype=torch.bfloat16).to("cuda")You can also load SmolVLM with 4/8-bit quantization using bitsandbytes, torchao or Quanto. Refer to this page for other options.
python
from transformers import AutoModelForVision2Seq, BitsAndBytesConfigimport torchquantization_config = BitsAndBytesConfig(load_in_8bit=True)model = AutoModelForVision2Seq.from_pretrained( "HuggingFaceTB/SmolVLM-Base", quantization_config=quantization_config,)Vision Encoder Efficiency: Adjust the image resolution by setting size={"longest_edge": N*384} when initializing the processor, where N is your desired value. The default N=4 works well, which results in input images of
size 1536×1536. For documents, N=5 might be beneficial. Decreasing N can save GPU memory and is appropriate for lower-resolution images. This is also useful if you want to fine-tune on videos.
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:
SmolVLM is built upon the shape-optimized SigLIP as image encoder and SmolLM2 for text decoder part.
We release the SmolVLM checkpoints under the Apache 2.0 license.
The training data comes from The Cauldron and Docmatix datasets, with emphasis on document understanding (25%) and image captioning (18%), while maintaining balanced coverage across other crucial capabilities like visual reasoning, chart comprehension, and general instruction following.
| Model | MMMU (val) | MathVista (testmini) | MMStar (val) | DocVQA (test) | TextVQA (val) | Min GPU RAM required (GB) |
|---|---|---|---|---|---|---|
| SmolVLM | 38.8 | 44.6 | 42.1 | 81.6 | 72.7 | 5.02 |
| Qwen-VL 2B | 41.1 | 47.8 | 47.5 | 90.1 | 79.7 |
| 13.70 |
| InternVL2 2B | 34.3 | 46.3 | 49.8 | 86.9 | 73.4 | 10.52 |
| PaliGemma 3B 448px | 34.9 | 28.7 | 48.3 | 32.2 | 56.0 | 6.72 |
| moondream2 | 32.4 | 24.3 | 40.3 | 70.5 | 65.2 | 3.87 |
| MiniCPM-V-2 | 38.2 | 39.8 | 39.1 | 71.9 | 74.1 | 7.88 |
| MM1.5 1B | 35.8 | 37.2 | 0.0 | 81.0 | 72.5 | NaN |