SmolVLM: lightweight multimodal model for video, image, and text analysis, optimized for devices
10K+
Tiny LLM family built for speed, edge devices, and local development.
An OCI model artifact for the SmolLM2 Instruct family (135M · 360M · 1.7B), packaged as Safetensors with a vLLM-oriented layout/config for convenient distribution via Docker Hub.
*.safetensors)*-vllm images: These are provided today to support vLLM-compatible safetensor models. Their long-term role is still under discussion — options include continuing them as-is or exploring OCI index–based distribution (to offer --platform-like behavior). The approach may evolve depending on distribution standards and framework compatibility, with the goal of keeping things practical and developer-friendly.Content type
Model
Digest
sha256:4951f8c98…
Size
259.9 MB
Last updated
6 months ago
docker model pull ai/smollm2-vllmPulls:
388
Last week