LLM EXPLORER 59,420 MODELS INDEXED

Qwen3 32B MLX 6bit by Qwen

By Qwen · 154 downloads
HF Repo ↗ Copy URL

Qwen3 32B MLX 6bit is an open-source language model by Qwen. Features: 32b LLM, VRAM: 25.5GB, Context: 40K, License: apache-2.0, Quantized, LLM Explorer Score: 0.19.

  Arxiv:2309.00071   Arxiv:2505.09388   6-bit   6bit   Conversational   Mlx   Quantized   Qwen3   Region:us   Safetensors   Sharded   Tensorflow
Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen3-32B-MLX-6bit 

Qwen3 32B MLX 6bit Parameters and Internals

LLM NameQwen3 32B MLX 6bit
Repository 🤗https://huggingface.co/Qwen/Qwen3-32B-MLX-6bit 
Model Size32b
Required VRAM25.5 GB
Updated2026-07-14
MaintainerQwen
Model Typeqwen3
Model Files  5.4 GB: 1-of-5   5.3 GB: 2-of-5   5.3 GB: 3-of-5   5.3 GB: 4-of-5   4.2 GB: 5-of-5
Quantization Type6bit
Model ArchitectureQwen3ForCausalLM
Licenseapache-2.0
Context Length40960
Model Max Length40960
Transformers Version4.51.3
Tokenizer ClassQwen2Tokenizer
Padding Token<|endoftext|>
Vocabulary Size151936
Torch Data Typebfloat16
Errorsreplace

Best Alternatives to Qwen3 32B MLX 6bit

Best Alternatives
Context / RAM
Downloads
Likes
Qwen3 32B Bnb 4bit40K / 19.2 GB836417
... E5 Grad Accum 16 Merged 16bit40K / 65.8 GB280
...llcpt E5 Baseline Merged 16bit40K / 65.8 GB240
...line Merged 16bit Merged 16bit40K / 65.8 GB160
SERA 32B 6bit40K / 26.6 GB272
SERA 32B 4bit40K / 18.5 GB121
SERA 32B GA 4bit40K / 18.5 GB71
Qwen3 32B Unsloth Bnb 4bit40K / 39.5 GB667415
Qwen3 32B 4bit40K / 18.5 GB24366
Qwen3 32B MLX 8bit40K / 33.7 GB74912
Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen3-32B-MLX-6bit.