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)

Raw benchmark score — scale varies by task

Showing 19 models

#	Model	Domain	Task	Params ↑	GFLOPs	VRAM	Score	Speed	Arch	Source
1	CodeGen2-1B	Code	Code Generation	1.0B	—	2.3 GB	9.1HumanEval+	—	—	EvalPlus
2	StarCoderBase-1B	Code	Code Generation	1.0B	—	2.3 GB	12.2HumanEval+	—	—	EvalPlus
3	zyte-1B	Code	Code Generation	1.0B	—	2.3 GB	1.8HumanEval+	—	—	EvalPlus
4	SantaCoder-1.1B	Code	Code Generation	1.1B	—	2.6 GB	14.0HumanEval+	—	—	EvalPlus
5	DeepSeek-Coder-1.3B-base	Code	Code Generation	1.3B	—	3.0 GB	25.6HumanEval+	—	—	EvalPlus
6	DeepSeek-Coder-1.3B-instruct	Code	Code Generation	1.3B	—	3.0 GB	60.4HumanEval+	—	—	EvalPlus
7	InCoder-1.3B	Code	Code Generation	1.3B	—	3.0 GB	11.0HumanEval+	—	—	EvalPlus
8	CodeGen-2B	Code	Code Generation	2.0B	—	4.7 GB	22.6HumanEval+	—	—	EvalPlus
9	CodeT5+-2B	Code	Code Generation	2.0B	—	4.7 GB	22.0HumanEval+	—	—	EvalPlus
10	codegemma-2b	Code	Code Generation	2.0B	—	4.7 GB	20.7HumanEval+	—	—	EvalPlus
11	gemma-2b	Code	Code Generation	2.0B	—	4.7 GB	20.7HumanEval+	—	—	EvalPlus
12	gemma-2b-it	Code	Code Generation	2.0B	—	4.7 GB	15.2HumanEval+	—	—	EvalPlus
13	GPT-Neo-2.7B	Code	Code Generation	2.7B	—	6.3 GB	6.7HumanEval+	—	—	EvalPlus
14	PolyCoder-2.7B	Code	Code Generation	2.7B	—	6.3 GB	6.1HumanEval+	—	—	EvalPlus
15	phi-2-2.7B	Code	Code Generation	2.7B	—	6.3 GB	45.1HumanEval+	—	—	EvalPlus
16	CodeGen2-3B	Code	Code Generation	3.0B	—	7.0 GB	12.8HumanEval+	—	—	EvalPlus
17	StarCoder2-3B	Code	Code Generation	3.0B	—	7.0 GB	27.4HumanEval+	—	—	EvalPlus
18	StarCoderBase-3B	Code	Code Generation	3.0B	—	7.0 GB	15.9HumanEval+	—	—	EvalPlus
19	stable-code-3B	Code	Code Generation	3.0B	—	7.0 GB	25.6HumanEval+	—	—	EvalPlus