Qwen-Image
Qwen · Qwen/Qwen-Image
Weights license: Apache 2.0 · code and components may use other terms
Quality
Quality · LMArena · text-to-image
LMArena 2026-10-07 · human-preference rating · higher is better · this model outlined
- Color = publisher
Source: LMArena 2026-10-07 · collected 2026-10-25 · dataset · CC BY 4.0 · models without a rating are not drawn (no rating ≠ 0)
1,057 ± 3 · LMArena · text-to-image · Confidence interval 1,054–1,059 · 84,550 votes · Arena also lists qwen-image-prompt-extend, the same model behind a prompt-rewriting step; only the plain qwen-image row is used
Artificial Analysis is excluded: its free tier is internal use only.
Reported by the publisher
Quality × memory
No comparable quality × memory pair
Image, video and audio models have no shared quality scale, and their memory comes from the publisher's documented minimum path, so quality is not charted against memory.
Consider this checkpoint for text-to-image generation or editing when you need to evaluate its separate image and text components.
The published component count and repository size do not establish a minimum VRAM requirement.
Runs on
Enough memory 72
A publisher documents a path within this memory, on this platform (CUDA or Apple MPS). Conditional paths need offload or quantization.
Show all 72
Hardware profiles grouped by the publisher's documented minimum memory path and platform. Not a measured run.
Measured speed
No published measurement for this checkpoint
Generation time per image is shown only when a source publishes it with hardware, build and setup. It is never estimated.
How to run
NVIDIA GPU acceleration is recommended; no minimum is published in the model card. Resolution, precision and whether all conditioning components are resident change the requirement. Official hardware evidence
Documented ways to run it
Copied as printed in the linked source (read 2026-10-18); not run by this site.
pip install git+https://github.com/huggingface/diffusersfrom diffusers import DiffusionPipeline
import torch
model_name = "Qwen/Qwen-Image"
# Load the pipeline
if torch.cuda.is_available():
torch_dtype = torch.bfloat16
device = "cuda"
else:
torch_dtype = torch.float32
device = "cpu"
pipe = DiffusionPipeline.from_pretrained(model_name, torch_dtype=torch_dtype)
pipe = pipe.to(device)
positive_magic = {
"en": ", Ultra HD, 4K, cinematic composition.", # for english prompt
"zh": ", 超清,4K,电影级构图." # for chinese prompt
}
# Generate image
prompt = '''A coffee shop entrance features a chalkboard sign reading "Qwen Coffee 😊 $2 per cup," with a neon light beside it displaying "通义千问". Next to it hangs a poster showing a beautiful Chinese woman, and beneath the poster is written "π≈3.1415926-53589793-23846264-33832795-02384197". Ultra HD, 4K, cinematic composition'''
negative_prompt = " " # using an empty string if you do not have specific concept to remove
# Generate with different aspect ratios
aspect_ratios = {
"1:1": (1328, 1328),
"16:9": (1664, 928),
"9:16": (928, 1664),
"4:3": (1472, 1140),
"3:4": (1140, 1472),
"3:2": (1584, 1056),
"2:3": (1056, 1584),
}
width, height = aspect_ratios["16:9"]
image = pipe(
prompt=prompt + positive_magic["en"],
negative_prompt=negative_prompt,
width=width,
height=height,
num_inference_steps=50,
true_cfg_scale=4.0,
generator=torch.Generator(device="cuda").manual_seed(42)
).images[0]
image.save("example.png")Evidence level: Publisher documentation linked; no execution reproduced by this site.
Before downloading
- Check the exact checkpoint and license linked here.
- Confirm a compatible runtime and precision or quantization for your platform.
- Allow for context, batch size and any extra media components beyond the weights.
Technical details
What the source and the catalog store for this checkpoint. Missing fields are left out, never zero.
- Checkpoint / version
- Qwen/Qwen-Image
- Task and modality
- text-to-image / editing · image
- Count and basis
- 27.1B · manufacturer model card
- Released
- QwenLM/Qwen-Image README (GitHub), News section · read 2026-10-09source
- Published repository size
- 57.7 GBNot runtime memory or a minimum download.files
- Published components
- MMDiT · 20B · Qwen2.5-VL · 7B · VAE · 54M / 73M
- Runtime and precision
- No local execution verified.
- Commercial use
- Allowed Apache License 2.0 text (SPDX) · read 2026-10-18license text
The text-to-image checkpoint has separate text encoder and VAE components; repository size and transformer parameters do not alone establish an inference minimum.
Sources
Identity and published specifications come from the checkpoint page. Benchmark scores belong to the named release and do not measure speed on your computer.
- Publisher checkpoint and model card
- Weight license and use terms
- Quality benchmark source · Artificial Analysis image · excluded (free tier is internal use only)
- Publisher hardware documentation
Catalog reviewed: 2026-09-30.


















































