LLM EXPLORER 55,709 MODELS INDEXED

Llama 4 Maverick 17B 128E Instruct 6bit by mlx-community

By mlx-community · 125 downloads
HF Repo ↗ Copy URL

Llama 4 Maverick 17B 128E Instruct 6bit is an open-source language model by mlx-community. Features: 17b LLM, VRAM: 201.1GB, License: other, Quantized, Instruction-Based.

  6bit   Ar Base model:finetune:meta-llama... Base model:meta-llama/llama-4-...   Conversational   De   En   Endpoints compatible   Es   Facebook   Fr   Hi   Id   Image-text-to-text   Instruct   It   Llama   Llama-4   Llama4   Meta   Mlx   Pt   Pytorch   Quantized   Region:us   Safetensors   Sharded   Tensorflow   Th   Tl   Vi
Model Card on HF 🤗: https://huggingface.co/mlx-community/Llama-4-Maverick-17B-128E-Instruct-6bit 

Llama 4 Maverick 17B 128E Instruct 6bit Parameters and Internals

LLM NameLlama 4 Maverick 17B 128E Instruct 6bit
Repository 🤗https://huggingface.co/mlx-community/Llama-4-Maverick-17B-128E-Instruct-6bit 
Base Model(s)  meta-llama/Llama-4-Maverick-17B-128E   meta-llama/Llama-4-Maverick-17B-128E
Model Size17b
Required VRAM201.1 GB
Updated2026-08-08
Maintainermlx-community
Model Typellama4
Instruction-BasedYes
Model Files  1.9 GB: 1-of-73   4.4 GB: 2-of-73   4.4 GB: 3-of-73   4.8 GB: 4-of-73   4.4 GB: 5-of-73   4.4 GB: 6-of-73   4.8 GB: 7-of-73   4.4 GB: 8-of-73   4.4 GB: 9-of-73   4.8 GB: 10-of-73   4.4 GB: 11-of-73   4.4 GB: 12-of-73   4.8 GB: 13-of-73   4.4 GB: 14-of-73   4.4 GB: 15-of-73   4.8 GB: 16-of-73   4.4 GB: 17-of-73   4.4 GB: 18-of-73   4.8 GB: 19-of-73   4.4 GB: 20-of-73   4.4 GB: 21-of-73   4.8 GB: 22-of-73   4.4 GB: 23-of-73   4.4 GB: 24-of-73   4.8 GB: 25-of-73   4.4 GB: 26-of-73   4.4 GB: 27-of-73   4.8 GB: 28-of-73   4.4 GB: 29-of-73   4.4 GB: 30-of-73   4.8 GB: 31-of-73   4.4 GB: 32-of-73   4.4 GB: 33-of-73   4.8 GB: 34-of-73   4.4 GB: 35-of-73   4.4 GB: 36-of-73   4.8 GB: 37-of-73   4.4 GB: 38-of-73   4.4 GB: 39-of-73   4.8 GB: 40-of-73   4.4 GB: 41-of-73   4.4 GB: 42-of-73   4.8 GB: 43-of-73   4.4 GB: 44-of-73   4.4 GB: 45-of-73
Supported Languagesar de en es fr hi id it pt th tl vi
Quantization Type6bit
Model ArchitectureLlama4ForConditionalGeneration
Licenseother
Transformers Version4.51.0
Torch Data Typebfloat16

Best Alternatives to Llama 4 Maverick 17B 128E Instruct 6bit

Best Alternatives
Context / RAM
Downloads
Likes
...Maverick 17B 16E Instruct 4bit0K / 146.3 GB4197
...Maverick 17B 16E Instruct 6bit0K / 213.8 GB1312
... 4 Scout 17B 16E Instruct 4bit0K / 61 GB120312
... 4 Scout 17B 16E Instruct 6bit0K / 88.8 GB3335
... 4 Scout 17B 16E Instruct 8bit0K / 114.7 GB5884
...averick 17B 128E Instruct 4bit0K / 140.8 GB2042
Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Llama-4-Maverick-17B-128E-Instruct-6bit.