What are the hardware requirements for Qwen Image Bench?

Qwen Image Bench requires approximately 54.7 GB of VRAM. Quantized variants may run on less VRAM; see the Quantized Models section on this page.

Who developed Qwen Image Bench and how large is it?

Qwen Image Bench is developed by Qwen, a model with 27b 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 Qwen Image Bench?

Qwen Image Bench is hosted on Hugging Face and linked from this page. LLM Explorer also lists quantized variants and similar alternatives if available.

Qwen Image Bench by Qwen — VRAM 54.7GB

Name: Qwen Image Bench
Author: Qwen

Qwen Image Bench is an open-source language model by Qwen. Features: 27b LLM, VRAM: 54.7GB, License: apache-2.0, LLM Explorer Score: 0.33.

Arxiv:2605.28091 Base model:finetune:qwen/qwen3... Base model:qwen/qwen3.6-27b Benchmark Conversational Deploy:azure En Endpoints compatible Evaluation Image-text-to-text Judge-model Qwen Qwen3 5 Region:us Safetensors Sharded Tensorflow Text-to-image Zh

Model Card on HF 🤗: https://huggingface.co/Qwen/Qwen-Image-Bench

Qwen Image Bench Parameters and Internals

LLM Name	Qwen Image Bench
Repository 🤗	https://huggingface.co/Qwen/Qwen-Image-Bench
Base Model(s)	Qwen/Qwen3.6-27B Qwen/Qwen3.6-27B
Model Size	27b
Required VRAM	54.7 GB
Updated	2026-08-10
Maintainer	Qwen
Model Type	qwen3_5
Model Files	49.8 GB: 1-of-2 4.9 GB: 2-of-2
Supported Languages	en zh
Model Architecture	Qwen3_5ForConditionalGeneration
License	apache-2.0
Model Max Length	262144
Transformers Version	5.2.0
Tokenizer Class	TokenizersBackend
Padding Token	<\|endoftext\|>
Errors	replace

Rank the Qwen Image Bench 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 Qwen Image Bench

Best Alternatives	Context / RAM	Downloads	Likes
...wopus3.8 27B Flash Mlx OQ4e 1M	1024K / 16.9 GB	211	1
Agens Pilot	1024K / 54.7 GB	135	1
Qwen3.6 27B	0K / 55.4 GB	6728138	2197
Qwen3.8 27B	0K / 55.6 GB	2	8753
Qwen3.8 27B MLX	0K / 16 GB	1411	18
Qwen3.8 27B FP8	0K / 0.4 GB	123157	416
Qwen3.8 27B	0K / 55.6 GB	2727	21
Qwen3.6 27B NVFP4	0K / 22 GB	1887525	419
Qwen3.8 27B NVFP4	0K / 22 GB	19195	74
Qwen3.6 27B FP8	0K / 0.4 GB	7740472	342

Note: green Score (e.g. "73.2") means that the model is better than Qwen/Qwen-Image-Bench.