How to run Qwen-Image locally

Alibaba (Qwen) · Text → image · 20B params · Apache-2.0 (commercial OK)

Alibaba's 20B MMDiT foundation model, notable for state-of-the-art text rendering (especially Chinese) under a fully permissive Apache license.

What it costs to run — $0 markup

On your own machine — $0. Runs free if you have about 24 GB of VRAM (fp8/GGUF quant ~16-24GB; full bf16 ~40GB+ or offload), via ComfyUI / Diffusers.
Rent a GPU — from $0.49/hr. Fits RTX A6000 48GB at the direct vendor price ($0 markup) — pay only for the minutes you generate.
Download the weights — free. Open weights at Qwen/Qwen-Image.

Note: generative-media models are billed per image / per second / per minute on hosted services — not per token. Running locally or on your own rented GPU is usually far cheaper and keeps your data on your machine.

Key facts

Does	Text → image
VRAM to run	~24 GB (fp8/GGUF quant ~16-24GB; full bf16 ~40GB+ or offload)
Download	~41 GB
Parameters	20B
License	Apache-2.0 (commercial use OK)
Run with	ComfyUI / Diffusers

Get Qwen-Image on Hugging Face →

More image models

FLUX.1 [dev] — Text → image, ~12 GB VRAM
Stable Diffusion XL 1.0 (SDXL) — Text → image, ~8.0 GB VRAM
Stable Diffusion 1.5 — Text → image, ~4.0 GB VRAM
FLUX.1 Kontext [dev] — Image editing, ~12 GB VRAM
FLUX.1 [schnell] — Text → image, ~12 GB VRAM
Stable Diffusion 3.5 Large — Text → image, ~12 GB VRAM

Browse: all media models · chat / LLM models