SmolVLM Grows Smaller – Introducing the 256M & 500M Models!

We’re excited to announce two new additions to the SmolVLM family: SmolVLM-256M and SmolVLM-500M. That’s right—256M parameters, making it the smallest Vision Language Model in the world!

We built on everything we learned from SmolVLM 2B while focusing on efficiency, data mixtures, and new design trade-offs. We are excited to introduce a pair of models that preserve strong multimodal performance in a fraction of the footprint.

This release comes with four checkpoints: two base models and two instruction fine-tuned models with sizes 256M and 500M parameters. These models can be loadable directly to transformers, MLX and ONNX, and we have demos for transformers and WebGPU (with ONNX). You can find all the models and the demo for this release here .

With just 256 million parameters, this model stands as the tiniest VLM ever. Despite its small size, it packs a surprising punch. It’s more than capable on many multimodal tasks, including:

If you need more performance headroom while still keeping the memory usage low, SmolVLM-500M is our half-billion-parameter compromise. It’s significantly smaller than the previous 2B release yet manages to push scores on tasks like DocVQA and MMMU closer to the bigger models. We also found this model to be more robust to prompting, which makes it out-of-the-box better fitted for production. But both models do great when fine-tuned.

We found the performance gap wasn’t big enough to justify the heavier encoder for our 256M and 500M models. So, we decided to go small on the vision encoder, too. As a bonus, the smaller encoder processes images at a larger resolution, which (per research from Apple and Google ) can often yield better visual understanding without ballooning parameter counts.

We're now reaching model parity with the SmolLM2 family (135M, 360M, 1.7B), so you have a complete set of smaller LLM + VLM combos to play with.

When we released SmolVLM 2B, the community response was fantastic: The model is very light weight, open-source and permissive, and easy to integrate into existing workflows. But we wanted to push this approach even further for people working with constrained devices, consumer laptops, or even potentially browser-based inference. That’s where our new 256M and 500M models come in. On the other side, for people trying to process huge amounts of data, these models can run at a fraction of the cost of the 2B model.

In the last year, we trained two 80B VLMs and reduced them to 8B. Then for SmolVLM we took the challenge of reducing that 2B. And what we learned was that we could push the frontier way further! We are excited to show that at 256M and 500M we can still get great performance. Our new 256M model is the smallest VLM ever released, yet it surpasses the performance of our Idefics 80B model from just 17 months ago.