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)

Max Size≤10M ≤50M ≤100M ≤500M All

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

Score

Raw Context (0-100)

Percentile rank within each task — comparable across domains

Showing 14 models

#	Model	Domain	Task	Params ↑	GFLOPs	VRAM	Context	Speed	Arch	Source
1	NeuML/pubmedbert-base-embeddings-1M	Embedding	Text Embedding	1.00M	—	2 MB	—	—	dense	MTEB
2	NeuML/pubmedbert-base-embeddings-500K	Embedding	Text Embedding	1.00M	—	2 MB	—	—	dense	MTEB
3	axiotic/ogma-micro	Embedding	Text Embedding	2.00M	—	9 MB	—	—	dense	MTEB
4	minishlab/potion-base-2M	Embedding	Text Embedding	2.00M	—	7 MB	—	—	dense	MTEB
5	NeuML/pubmedbert-base-embeddings-2M	Embedding	Text Embedding	2.00M	—	7 MB	—	—	dense	MTEB
6	axiotic/ogma-mini	Embedding	Text Embedding	4.00M	—	13 MB	—	—	dense	MTEB
7	minishlab/potion-base-4M	Embedding	Text Embedding	4.00M	—	14 MB	—	—	dense	MTEB
8	cross-encoder/ms-marco-TinyBERT-L2-v2	Embedding	Text Embedding	4.00M	—	17 MB	—	—	cross-encoder	MTEB
9	google/yamnet	Embedding	Text Embedding	4.00M	—	14 MB	—	—	dense	MTEB
10	minishlab/M2V_base_output	Embedding	Text Embedding	8.00M	—	29 MB	—	—	dense	MTEB
11	minishlab/potion-base-8M	Embedding	Text Embedding	8.00M	—	29 MB	—	—	dense	MTEB
12	NeuML/pubmedbert-base-embeddings-8M	Embedding	Text Embedding	8.00M	—	30 MB	—	—	dense	MTEB
13	axiotic/ogma-small	Embedding	Text Embedding	9.00M	—	33 MB	—	—	dense	MTEB
14	marksverdhei/Qwen3-Voice-Embedding-12Hz-0.6B	Embedding	Text Embedding	9.00M	—	17 MB	—	—	dense	MTEB