Opening article
SmolVLM2 is a new family of video‑understanding models released by Hugging Face that prioritises efficiency over size, offering 2.2B, 500M and 256M parameter variants that can run on devices from phones to servers. The models are MLX ready with Python and Swift APIs and are available for inference via a conversational API or dedicated scripts. The 2.2B model leads 2B‑range models on the Video‑MME benchmark, while the 500M and 256M models claim to be the smallest video language models yet released, outperforming larger models per memory consumption. Hugging Face provides tutorials, fine‑tuning notebooks, and an interactive chat interface for testing the models.
SmolVLM2 represents a fundamental shift in how we think about video understanding - moving from massive models that require substantial computing resources to efficient models that can run anywhere. Our goal is simple: make video understanding accessible across all devices and use cases, from phones to servers.
We are releasing models in three sizes (2.2B, 500M and 256M), MLX ready (Python and Swift APIs) from day zero. We've made all models and demos available in this collection .
Want to try SmolVLM2 right away? Check out our interactive chat interface where you can test visual and video understanding capabilities of SmolVLM2 2.2B through a simple, intuitive interface.
We are introducing three new models with 256M, 500M and 2.2B parameters. The 2.2B model is the go-to choice for vision and video tasks, while the 500M and 256M models represent the smallest video language models ever released .
Our new SmolVLM2-500M-Video-Instruct model has video capabilities very close to SmolVLM 2.2B, but at a fraction of the size: we're getting the same video understanding capabilities with less than a quarter of the parameters 🤯.
While they're small in size, they outperform any existing models per memory consumption. Looking at Video-MME (the go-to scientific benchmark in video), SmolVLM2 joins frontier model families on the 2B range and we lead the pack in the even smaller space.
Video-MME stands out as a comprehensive benchmark due to its extensive coverage across diverse video types, varying durations (11 seconds to 1 hour), multiple data modalities (including subtitles and audio), and high-quality expert annotations spanning 900 videos totaling 254 hours. Learn more here .
Compared with the previous SmolVLM family, our new 2.2B model got better at solving math problems with images, reading text in photos, understanding complex diagrams, and tackling scientific visual questions. This shows in the model performance across different benchmarks:
Efficient video models enable broader deployment of multimodal AI, reducing hardware barriers and fostering innovation across mobile and edge applications.