LLM EXPLORER 66,613 MODELS INDEXED

Mistral Large 4 1T A52B MLX Q3i by inferencerlabs

By inferencerlabs · 1397 downloads
HF Repo ↗ Copy URL

Mistral Large 4 1T A52B MLX Q3i is an open-source language model by inferencerlabs. Features: LLM, VRAM: 175.3GB.

Base model:mistralai/mistral-l... Base model:quantized:mistralai...   Conversational   En   Image-text-to-text   Mistral large4   Mlx   Quantized   Region:us   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/inferencerlabs/Mistral-Large-4-1T-A52B-MLX-Q3i 

Mistral Large 4 1T A52B MLX Q3i Parameters and Internals

LLM NameMistral Large 4 1T A52B MLX Q3i
Repository 🤗https://huggingface.co/inferencerlabs/Mistral-Large-4-1T-A52B-MLX-Q3i 
Base Model(s)  Mistral Large 4 1T A52B   mistralai/Mistral-Large-4-1T-A52B
Required VRAM175.3 GB
Updated2026-11-04
Maintainerinferencerlabs
Model Typemistral_large4
Model Files  3.9 GB: 1-of-108   3.9 GB: 2-of-108   4.2 GB: 3-of-108   4.2 GB: 4-of-108   3.9 GB: 5-of-108   4.2 GB: 6-of-108   4.2 GB: 7-of-108   3.9 GB: 8-of-108   4.2 GB: 9-of-108   4.2 GB: 10-of-108   3.9 GB: 11-of-108   4.2 GB: 12-of-108   4.2 GB: 13-of-108   3.9 GB: 14-of-108   4.2 GB: 15-of-108   4.2 GB: 16-of-108   3.9 GB: 17-of-108   4.2 GB: 18-of-108   4.8 GB: 19-of-108   3.9 GB: 20-of-108   4.7 GB: 21-of-108   4.2 GB: 22-of-108   3.9 GB: 23-of-108   4.2 GB: 24-of-108   4.2 GB: 25-of-108   3.9 GB: 26-of-108   4.2 GB: 27-of-108   4.2 GB: 28-of-108   3.9 GB: 29-of-108   4.2 GB: 30-of-108   4.8 GB: 31-of-108   3.9 GB: 32-of-108   4.8 GB: 33-of-108   4.3 GB: 34-of-108   3.9 GB: 35-of-108   4.2 GB: 36-of-108   4.2 GB: 37-of-108   3.9 GB: 38-of-108   4.6 GB: 39-of-108   4.7 GB: 40-of-108   3.9 GB: 41-of-108   4.3 GB: 42-of-108
Supported Languagesen
Model ArchitectureMistralLarge4ForConditionalGeneration
Transformers Version5.21.0.dev0

Best Alternatives to Mistral Large 4 1T A52B MLX Q3i

Best Alternatives
Context / RAM
Downloads
Likes
Mistral Large 4 1T A52B0K / 880 GB29451398
Mistral Large 4 1T A52B FP80K / 880 GB1455
Note: green Score (e.g. "73.2") means that the model is better than inferencerlabs/Mistral-Large-4-1T-A52B-MLX-Q3i.