Small Model Hub

3,618 efficient AI models across 9 domains · Charts · API

Domain

All (3,618)Vision (1,615)3D (743)Embedding (472)Multimodal (305)Edge/Mobile (172)Code (125)Speech (89)Video (49)NLP (48)

Task

All Tasks Vision-Language (285)UI Understanding (14)image to text (2)robotics (2)Image Classification (2)

Max Size≤10M ≤50M ≤100M ≤500M All

Max VRAM≤512MB ≤1GB ≤2GB ≤4GB ≤8GB All

Score

Raw Context (0-100)

Raw benchmark score — scale varies by task

Showing 14 models

#	Model	Domain	Task	Params ↑	GFLOPs	VRAM	Score	Speed	Arch	Source
1	OpenAI-Clip	Multimodal	Image Classification	150.0M	—	360 MB	——	—	—	Qualcomm AI Hub
2	SmolVLM-256M	Multimodal	Vision-Language	260.0M	—	624 MB	41.7MMBench	—	—	OpenVLM
3	SmolVLM2-256M	Multimodal	Vision-Language	260.0M	—	624 MB	11.9MMBench	—	—	OpenVLM
4	SmolVLM-500M	Multimodal	Vision-Language	510.0M	—	1.2 GB	53.8MMBench	—	—	OpenVLM
5	SmolVLM2-500M	Multimodal	Vision-Language	510.0M	—	1.2 GB	53.1MMBench	—	—	OpenVLM
6	H2OVL-800M	Multimodal	Vision-Language	830.0M	—	1.9 GB	56.3MMBench	—	—	OpenVLM
7	InternVL3-1B	Multimodal	Vision-Language	940.0M	—	2.2 GB	72.3MMBench	—	—	OpenVLM
8	InternVL2-1B	Multimodal	Vision-Language	1.0B	—	2.3 GB	65.2MMBench	—	—	OpenVLM
9	LLaVA-OneVision-0.5B	Multimodal	Vision-Language	1.0B	—	2.3 GB	61.6MMBench	—	—	OpenVLM
10	LLaVA-OneVision-0.5B (SI)	Multimodal	Vision-Language	1.0B	—	2.3 GB	55.8MMBench	—	—	OpenVLM
11	InternVL2.5-1B	Multimodal	Vision-Language	1.0B	—	2.3 GB	——	—	—	OpenVLM
12	InternVL2.5-1B-MPO	Multimodal	Vision-Language	1.0B	—	2.3 GB	70.1MMBench	—	—	OpenVLM
13	Ovis2-1B	Multimodal	Vision-Language	1.3B	—	3.0 GB	71.6MMBench	—	—	OpenVLM
14	Kosmos2	Multimodal	Vision-Language	1.7B	—	4.0 GB	1.1MMBench	—	—	OpenVLM