What are the hardware requirements for Qwen3 235B A22B Thinking 2507 3bit DWQ?

Qwen3 235B A22B Thinking 2507 3bit DWQ requires approximately 103 GB of VRAM and supports a context window of 256K tokens. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen3 235B A22B Thinking 2507 3bit DWQ and how large is it?

Qwen3 235B A22B Thinking 2507 3bit DWQ is developed by mlx-community, a model with 235b parameters. The model is published as open weights on Hugging Face and indexed on LLM Explorer with full benchmark history.

Where can I download or evaluate Qwen3 235B A22B Thinking 2507 3bit DWQ?

Qwen3 235B A22B Thinking 2507 3bit DWQ is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen3 235B A22B Thinking 2507 3bit DWQ by mlx-community — VRAM 103GB, 256K context

Name: Qwen3 235B A22B Thinking 2507 3bit DWQ
Author: mlx-community

Qwen3 235B A22B Thinking 2507 3bit DWQ is an open-source language model by mlx-community. Features: 235b LLM, VRAM: 103GB, Context: 256K, License: apache-2.0, Quantized, LLM Explorer Score: 0.19.

3-bit 3bit Base model:quantized:qwen/qwen... Base model:qwen/qwen3-235b-a22... Conversational Mlx Quantized Qwen3 moe Region:us Safetensors Sharded Tensorflow

Model Card on HF 🤗: https://huggingface.co/mlx-community/Qwen3-235B-A22B-Thinking-2507-3bit-DWQ

Qwen3 235B A22B Thinking 2507 3bit DWQ Parameters and Internals

LLM Name	Qwen3 235B A22B Thinking 2507 3bit DWQ
Repository 🤗	https://huggingface.co/mlx-community/Qwen3-235B-A22B-Thinking-2507-3bit-DWQ
Base Model(s)	Qwen3 235B A22B Thinking 2507 Qwen/Qwen3-235B-A22B-Thinking-2507
Model Size	235b
Required VRAM	103 GB
Updated	2026-07-08
Maintainer	mlx-community
Model Type	qwen3_moe
Model Files	5.4 GB: 1-of-20 5.1 GB: 2-of-20 5.1 GB: 3-of-20 5.4 GB: 4-of-20 5.1 GB: 5-of-20 5.1 GB: 6-of-20 5.1 GB: 7-of-20 5.4 GB: 8-of-20 5.1 GB: 9-of-20 5.1 GB: 10-of-20 5.1 GB: 11-of-20 5.4 GB: 12-of-20 5.1 GB: 13-of-20 5.1 GB: 14-of-20 5.1 GB: 15-of-20 5.4 GB: 16-of-20 5.1 GB: 17-of-20 5.1 GB: 18-of-20 5.1 GB: 19-of-20 4.6 GB: 20-of-20
Quantization Type	3bit
Model Architecture	Qwen3MoeForCausalLM
License	apache-2.0
Context Length	262144
Model Max Length	262144
Transformers Version	4.51.0
Tokenizer Class	Qwen2Tokenizer
Padding Token	<\|endoftext\|>
Vocabulary Size	151936
Torch Data Type	bfloat16
Errors	replace

Rank the Qwen3 235B A22B Thinking 2507 3bit DWQ capabilities

Have you tried this model? Rate its performance — this feedback helps the ML community find the right model for their needs.

Instruction Following and Task Automation
Factuality and Completeness of Knowledge
Censorship and Alignment
Data Analysis and Insight Generation
Text Generation
Text Summarization and Feature Extraction
Code Generation
Multi-Language Support and Translation

Best Alternatives to Qwen3 235B A22B Thinking 2507 3bit DWQ

Best Alternatives	Context / RAM	Downloads	Likes
...3 235B A22B Instruct 2507 4bit	256K / 131.4 GB	2658	3
...3 235B A22B Instruct 2507 8bit	256K / 227.7 GB	2591	2
...3 235B A22B Thinking 2507 8bit	256K / 227.7 GB	515	4
...5B A22B Instruct 2507 3bit DWQ	256K / 103 GB	174	5
...5B A22B Thinking 2507 4bit DWQ	256K / 131.4 GB	65	4
...5B A22B Instruct 2507 4bit DWQ	256K / 131.4 GB	66	2
Qwen3 235B A22B 4bit	40K / 131.4 GB	2526	9
Qwen3 235B A22B 8bit	40K / 238.3 GB	2042	4
Qwen3 235B A22B MLX 4bit	40K / 125.2 GB	399	16
Qwen3 235B A22B MLX 8bit	40K / 229.3 GB	168	9

Note: green Score (e.g. "73.2") means that the model is better than mlx-community/Qwen3-235B-A22B-Thinking-2507-3bit-DWQ.