AI Model Rankings

Best models for every use case, ranked by quality and sorted by minimum VRAM. Find the best model your hardware can run.

💬

Chat & General

General-purpose language models for conversation, writing, and reasoning

#	Model	Author	Params	Min VRAM	Action
1	SmolLM2 135M	HuggingFace	0.135B	0.3GB	Check
2	SmolLM2 360M	HuggingFace	0.36B	0.5GB	Check
3	Danube 3 500M	H2O.ai	0.5B	0.6GB	Check
4	Qwen 2.5 0.5B	Alibaba	0.5B	0.7GB	Check
5	TinyLlama 1.1B	TinyLlama	1.1B	1.2GB	Check
6	Gemma 3 1B	Google	1B	1.2GB	Check
7	Falcon 3 1B	TII	1B	1.2GB	Check
8	Llama 3.2 1B Instruct	Meta	1.24B	1.3GB	Check
9	Qwen 2.5 1.5B	Alibaba	1.5B	1.5GB	Check
10	DeepSeek R1 Distill 1.5B	DeepSeek	1.5B	1.5GB	Check
11	SmolLM2 1.7B	HuggingFace	1.7B	1.6GB	Check
12	Granite 3.3 2B	IBM	2B	1.8GB	Check
13	EXAONE 3.5 2.4B	LG AI	2.4B	2GB	Check
14	StableLM Zephyr 3B	Stability AI	3B	2.3GB	Check
15	Rocket 3B	Pansophic	3B	2.3GB	Check
16	Gemma 2 2B	Google	2.6B	2.3GB	Check
17	Qwen 2.5 3B	Alibaba	3B	2.5GB	Check
18	Falcon 3 3B	TII	3B	2.5GB	Check
19	Llama 3.2 3B Instruct	Meta	3.2B	2.6GB	Check
20	Nemotron Mini 4B	NVIDIA	4B	3GB	Check
21	Danube 3 4B	H2O.ai	4B	3GB	Check
22	Phi-3.5 Mini 3.8B	Microsoft	3.8B	3GB	Check
23	Phi-4 Mini 3.8B	Microsoft	3.8B	3GB	Check
24	Gemma 3 4B	Google	4B	3.2GB	Check
25	Yi 1.5 6B Chat	01.AI	6B	4.3GB	Check
26	Mistral 7B Instruct v0.3	Mistral AI	7.3B	5GB	Check
27	Falcon 3 7B	TII	7B	5GB	Check
28	OLMo 2 7B	Allen AI	7B	5GB	Check
29	OpenChat 3.5 7B	OpenChat	7B	5GB	Check
30	InternLM 2.5 7B	Shanghai AI Lab	7.7B	5.3GB	Check
31	Qwen 2.5 7B Instruct	Alibaba	7.6B	5.3GB	Check
32	DeepSeek R1 Distill 8B	DeepSeek	8B	5.5GB	Check
33	Llama 3.1 8B Instruct	Meta	8B	5.5GB	Check
34	Granite 3.3 8B	IBM	8B	5.5GB	Check
35	EXAONE 3.5 7.8B	LG AI	7.8B	5.5GB	Check
36	Gemma 2 9B Instruct	Google	9.2B	6.2GB	Check
37	Yi 1.5 9B Chat	01.AI	9B	6.2GB	Check
38	Falcon 3 10B	TII	10B	7GB	Check
39	Solar 10.7B	Upstage	10.7B	7.2GB	Check
40	Gemma 3 12B	Google	12B	8GB	Check
41	Mistral Nemo 12B	Mistral AI	12B	8GB	Check
42	Phi-4	Microsoft	14B	9.5GB	Check
43	Qwen 2.5 14B	Alibaba	14B	10GB	Check
44	Mistral Small 22B	Mistral AI	22B	14.5GB	Check
45	Gemma 3 27B	Google	27B	17.5GB	Check
46	Qwen 2.5 32B	Alibaba	32B	20GB	Check
47	Llama 3.1 70B Instruct	Meta	70B	42GB	Check

💻

Coding

Specialized models for code generation, completion, and debugging

#	Model	Author	Params	Min VRAM	Action
1	Qwen 2.5 Coder 0.5B	Alibaba	0.5B	0.9GB	Check
2	DeepSeek Coder 1.3B	DeepSeek	1.3B	1.3GB	Check
3	Qwen 2.5 Coder 1.5B	Alibaba	1.5B	1.5GB	Check
4	Yi Coder 1.5B	01.AI	1.5B	1.5GB	Check
5	CodeGemma 2B	Google	2B	1.9GB	Check
6	Stable Code 3B	Stability AI	3B	2.3GB	Check
7	StarCoder2 3B	BigCode	3B	2.4GB	Check
8	Qwen 2.5 Coder 3B	Alibaba	3B	2.5GB	Check
9	Code Llama 7B	Meta	7B	4.7GB	Check
10	DeepSeek Coder 6.7B	DeepSeek	6.7B	4.7GB	Check
11	StarCoder2 7B	BigCode	7B	5GB	Check
12	Qwen 2.5 Coder 7B	Alibaba	7.6B	5.3GB	Check
13	CodeGemma 7B	Google	8.5B	5.6GB	Check
14	Yi Coder 9B	01.AI	9B	6.2GB	Check
15	Code Llama 13B Instruct	Meta	13B	8.7GB	Check
16	Qwen 2.5 Coder 14B	Alibaba	14B	10GB	Check

🎨

Image Generation

Text-to-image models for art, photos, and design

#	Model	Author	Params	Min VRAM	Action
1	Stable Diffusion 2.1 Base (CoreML)	Stability AI / Apple	0.86B	2GB	Check
2	Stable Diffusion 1.5 (CoreML)	Runway	0.86B	2.5GB	Check
3	Stable Diffusion 1.5 (GGUF)	Runway / GPUStack	0.86B	2.5GB	Check
4	Stable Diffusion 2.1 (GGUF)	Stability AI	0.86B	2.5GB	Check
5	Stable Diffusion XL (CoreML)	Stability AI	3.5B	5GB	Check
6	SDXL Turbo (GGUF)	Stability AI	3.5B	5GB	Check
7	Stable Diffusion 3 Medium (GGUF)	Stability AI	2.5B	5.5GB	Check
8	FLUX.1 Schnell (GGUF)	Black Forest Labs	12B	14GB	Check
9	FLUX.1 Dev (GGUF)	Black Forest Labs	12B	14GB	Check

🎤

Speech-to-Text

Transcription and speech recognition models

#	Model	Author	Params	Min VRAM	Action
1	Whisper Tiny English (Quantized)	OpenAI	0.039B	0.1GB	Check
2	Whisper Tiny	OpenAI	0.039B	0.2GB	Check
3	Whisper Base	OpenAI	0.074B	0.3GB	Check
4	Whisper Base English	OpenAI	0.074B	0.3GB	Check
5	Whisper Small	OpenAI	0.24B	0.6GB	Check
6	Distil-Whisper Large v3	HuggingFace	0.76B	1GB	Check
7	Whisper Medium	OpenAI	0.77B	1.8GB	Check
8	Whisper Large v3 Turbo	OpenAI	0.81B	1.9GB	Check
9	Whisper Large v3	OpenAI	1.55B	3.5GB	Check

🔊

Text-to-Speech

Voice synthesis and text-to-speech models

#	Model	Author	Params	Min VRAM	Action
1	Piper TTS - Amy (English)	Rhasspy	0.02B	0.15GB	Check
2	Piper TTS - Lessac (English)	Rhasspy	0.02B	0.15GB	Check
3	Piper TTS - LibriTTS-R (English)	Rhasspy	0.02B	0.15GB	Check
4	Piper TTS - Spanish (MLS)	Rhasspy	0.02B	0.15GB	Check
5	Piper TTS - French (Siwis)	Rhasspy	0.02B	0.15GB	Check
6	Piper TTS - German (Thorsten)	Rhasspy	0.02B	0.15GB	Check
7	Piper TTS - Chinese (Huayan)	Rhasspy	0.02B	0.15GB	Check
8	Piper TTS - Japanese (Kokoro)	Rhasspy	0.02B	0.15GB	Check
9	Piper TTS - Korean	Rhasspy	0.02B	0.15GB	Check
10	Piper TTS - Russian (Irina)	Rhasspy	0.02B	0.15GB	Check
11	Piper TTS - Portuguese (Faber)	Rhasspy	0.02B	0.15GB	Check
12	Piper TTS - Italian (Riccardo)	Rhasspy	0.02B	0.15GB	Check
13	Piper TTS - Arabic (Kareem)	Rhasspy	0.02B	0.15GB	Check
14	Kokoro 82M TTS	Kokoro	0.082B	0.2GB	Check

🎵

Audio Generation

AI music and audio creation

#	Model	Author	Params	Min VRAM	Action
1	MusicGen Small	Meta	0.3B	1.5GB	Check

👁️

Multimodal / Vision

Models that understand both images and text

#	Model	Author	Params	Min VRAM	Action
1	Moondream 2	Moondream	1.8B	1.5GB	Check
2	Qwen2-VL 2B	Alibaba	2.2B	2GB	Check
3	MiniCPM-V 2.6	OpenBMB	2B	2.1GB	Check
4	PaliGemma 3B	Google	3B	2.5GB	Check
5	Phi-3.5 Vision	Microsoft	4.2B	3.2GB	Check
6	LLaVA 1.6 7B	LLaVA	7B	5GB	Check

🔗

Embedding

Text embedding models for search and retrieval

#	Model	Author	Params	Min VRAM	Action
1	BGE Small EN v1.5	BAAI	0.033B	0.1GB	Check
2	Snowflake Arctic Embed S	Snowflake	0.033B	0.1GB	Check
3	all-MiniLM-L6-v2	Sentence Transformers	0.023B	0.1GB	Check
4	Nomic Embed Text v1.5	Nomic AI	0.137B	0.3GB	Check
5	BGE Large EN v1.5	BAAI	0.335B	0.5GB	Check

Can't Run the Model You Want?

Cloud GPUs give you instant access to any model, any size.

RunPod — from $0.25/hr Vast.ai — from $0.15/hr

Need a GPU upgrade?

See our Best GPU for AI buying guide