AI Model Database

AI Models

Every model with its VRAM footprint, quantization options, and the hardware that can actually run it.

65 models

SEmbeddings

all-MiniLM-L6-v2

256 MB VRAM23M
NEmbeddings

nomic-embed-text v1.5

410 MB VRAM137M
MetaChat

Llama 3.2 1B Instruct

1.1 GB VRAM1.24B
OpenAITranscription

Whisper Large V3 Turbo

1.6 GB VRAM810M
OpenAITranscription

Whisper Large V3

1.5 GB VRAM1.55B
MetaChat

Llama 3.2 3B Instruct

2.8 GB VRAM3.21B
MicrosoftChat

Phi-4 Mini

3.3 GB VRAM3.82B
MicrosoftChat

Phi-3 Mini 3.8B Instruct

3 GB VRAM3.82B
GoogleChat

Gemma 3 4B

3.8 GB VRAM4.3B
SAImage gen

Stable Diffusion 3 Medium

5 GB VRAM2B
GoogleChat

Gemma 4 E2B

4.5 GB VRAM5.1B
SAImage gen

Stable Diffusion XL 1.0

6.5 GB VRAM6.6B
Chat

Arcee Trinity Nano 6B

5.4 GB VRAM6B
MistralChat

Mistral 7B Instruct v0.3

5.3 GB VRAM7.24B
DeepSeekReasoning

DeepSeek R1 Distill Qwen 7B

6.6 GB VRAM7.62B
QwenChat

Qwen 2.5 7B Instruct

5.5 GB VRAM7.62B
QwenCoding

Qwen 2.5 Coder 7B Instruct

6.6 GB VRAM7.62B
IChat

InternLM 2.5 7B Chat

6.7 GB VRAM7.74B
MetaChat

Llama 3.1 8B Instruct

6.7 GB VRAM8.03B
GoogleChat

Gemma 4 E4B

7 GB VRAM8B
MetaChat

LLaVA 1.6 13B

9.1 GB VRAM13B
QwenChat

Qwen3-8B Instruct

6.5 GB VRAM8.2B
GoogleChat

Gemma 2 9B Instruct

6.6 GB VRAM9.24B
Chat

GigaChat Lightning 10B

6 GB VRAM10B
SAImage gen

Stable Diffusion 3.5 Large

12.5 GB VRAM8.1B
MetaChat

Llama 3.2 11B Vision

10 GB VRAM11B
GoogleChat

Gemma 3 12B

10.5 GB VRAM12.2B
QwenChat

Qwen3-14B Instruct

10 GB VRAM14B
MistralCoding

Codestral 22B

15.1 GB VRAM22.2B
MicrosoftChat

Phi-3 Medium 14B Instruct

9.7 GB VRAM14B
MicrosoftReasoning

Phi-4 14B

12.6 GB VRAM14.7B
QwenChat

Qwen 2.5 14B Instruct

12.7 GB VRAM14.77B
MicrosoftCoding

StarCoder 2 15B

10.7 GB VRAM15.5B
DeepSeekCoding

DeepSeek Coder V2 Lite 16B

10.9 GB VRAM15.7B
GoogleChat

Gemma 2 27B Instruct

18.5 GB VRAM27.23B
NChat

Nemotron-Labs Diffusion 8B

19 GB VRAM8.5B
MetaCoding

Code Llama 34B Instruct

22.7 GB VRAM33.7B
BFLImage gen

FLUX.1 Dev

13 GB VRAM12B
MistralChat

Mistral Small 24B Instruct

20.5 GB VRAM24B
GoogleChat

Gemma 4 26B-A4B

24 GB VRAM25.2B
QwenChat

Qwen3.5-27B

19 GB VRAM27B
Chat

Arcee Trinity Mini 26B

20 GB VRAM26B
GoogleChat

Gemma 3 27B

22.3 GB VRAM27.23B
QwenChat

Qwen3.6-27B

20 GB VRAM27B
MistralChat

Mixtral 8x7B Instruct

31.4 GB VRAM46.7B
GoogleChat

Gemma 4 31B

28 GB VRAM30.7B
QwenChat

Qwen3-30B-A3B

23 GB VRAM30B
QwenCoding

Qwen 2.5 Coder 32B Instruct

21.9 GB VRAM32.5B
QwenReasoning

QwQ 32B Preview

21.9 GB VRAM32.5B
DeepSeekReasoning

DeepSeek R1 Distill Qwen 32B

28 GB VRAM32.5B
QwenChat

Qwen3-32B Instruct

25 GB VRAM32B
QwenChat

Qwen3.6-35B-A3B

25 GB VRAM35B
0Chat

Yi 1.5 34B Chat

29.5 GB VRAM34.4B
CohereChat

Command R 35B

30 GB VRAM35B
QwenChat

Qwen 2.5 72B Instruct

40.5 GB VRAM72.7B
MetaChat

Llama 3.1 70B Instruct

47 GB VRAM70.6B
QwenChat

Qwen3.5-122B-A10B

42 GB VRAM122B
NChat

NVIDIA Nemotron-3-super-120B-A12B

70 GB VRAM120B
MetaChat

Llama 3.3 70B Instruct

61 GB VRAM70.6B
MetaChat

Llama 4 Scout

75 GB VRAM109B
MistralChat

Mistral Large 2 123B

95 GB VRAM123B
QwenChat

Qwen3.5-397B (MoE)

230 GB VRAM397B
Chat

Arcee Trinity Large Thinking 400B

295 GB VRAM399B
DeepSeekReasoning

DeepSeek R1

360 GB VRAM671B
DeepSeekChat

DeepSeek V3

360 GB VRAM671B