# SmolVLM Grows Smaller - Introducing the 256M & 500M Models!

DevFeed: [SmolVLM Grows Smaller - Introducing the 256M & 500M Models!](<https://devfeed.tech/articles/smolvlm-grows-smaller-introducing-the-256m-500m-models-7480.md>)

Original publisher: [Read original article](<https://huggingface.co/blog/smolervlm>)

Author: Andres Marafioti; Miquel Farré; merve

Published: 2025-01-23T00:00:00Z

Content type: release

Language: en

Sources: [Hugging Face - Blog](<https://devfeed.tech/sources/hugging-face-blog.md>)

Topics: [vlm](<https://devfeed.tech/topics/vlm.md>), [releases](<https://devfeed.tech/topics/releases.md>)

Tags: [artificial-intelligence](<https://devfeed.tech/tags/artificial-intelligence.md>), [benchmarks](<https://devfeed.tech/tags/benchmarks.md>), [browser](<https://devfeed.tech/tags/browser.md>), [community](<https://devfeed.tech/tags/community.md>), [cost](<https://devfeed.tech/tags/cost.md>), [demo](<https://devfeed.tech/tags/demo.md>), [devices](<https://devfeed.tech/tags/devices.md>), [images](<https://devfeed.tech/tags/images.md>), [inference](<https://devfeed.tech/tags/inference.md>), [llm](<https://devfeed.tech/tags/llm.md>), [mlx](<https://devfeed.tech/tags/mlx.md>), [models](<https://devfeed.tech/tags/models.md>), [multimodal](<https://devfeed.tech/tags/multimodal.md>), [on-device](<https://devfeed.tech/tags/on-device.md>), [open-source](<https://devfeed.tech/tags/open-source.md>), [optimization](<https://devfeed.tech/tags/optimization.md>), [performance](<https://devfeed.tech/tags/performance.md>), [release](<https://devfeed.tech/tags/release.md>), [research](<https://devfeed.tech/tags/research.md>), [smolvlm](<https://devfeed.tech/tags/smolvlm.md>), [tokenization](<https://devfeed.tech/tags/tokenization.md>), [training](<https://devfeed.tech/tags/training.md>), [vision](<https://devfeed.tech/tags/vision.md>), [vlm](<https://devfeed.tech/tags/vlm.md>)

## AI overview

Hugging Face announces 256M and 500M SmolVLM vision-language models, including base and instruction-tuned checkpoints. The release emphasizes small-footprint multimodal performance, a smaller vision encoder, larger image resolution, and support for Transformers, MLX, ONNX, and WebGPU demos.

## Source excerpt

We're on a journey to advance and democratize artificial intelligence through open source and open science.