Qwen Image 3. Crisp text and fine detail, down to 10px.
Qwen's unified model for generating and editing images renders text and details as small as 10 pixels. It comes in two tiers, Qwen Image 3 and Qwen Image 3 Pro, at 1K or 2K, in formats from 1:4 to 4:1.
- Text and image input
- 1K or 2K output
- 13 aspect ratios, 1:4 to 4:1
- Up to 4 reference images
Qwen Image 3 is a unified image generation and editing model from Qwen, Alibaba's model family. One model takes a prompt, or a prompt with reference images, and it is built for precision: text and details as small as 10 pixels render cleanly, so menus, labels, and small print hold up in the frame. It ships in two tiers, Qwen Image 3 and Qwen Image 3 Pro, with the same inputs and settings.
On Treza both sit in the Image Generation node's model picker. Pick 1K or 2K and one of thirteen aspect ratios, including 1:4 and 4:1 strips for tall infographics, banners, and page headers. Up to 4 reference images go into one node for edits and compositions, and a finished still wires straight into the rest of the pipeline, which publishes as one API.
Text down to 10 pixels
Small print, price tags, and menu lines stay legible, with text and details rendered as small as 10 pixels.
Generate and edit in one model
Write a prompt to start fresh, or attach an image and describe the change. The same node does both.
Tall and wide formats
Thirteen aspect ratios from 1:4 to 4:1, for vertical infographics, banner strips, and everything in between.
Up to 4 references
Combine a product, a setting, and a style reference in one request and compose a single image from them.
What to make with Qwen Image 3
Prompts to try, and what to expect back. Each runs on an Image Generation node with Qwen Image 3 or Qwen Image 3 Pro picked in the model menu.
“A cafe menu board with six drinks and their prices, chalk lettering, 3:4”
A menu where every line and price reads clearly.
“A 1:4 vertical infographic on how coffee is roasted, five labeled steps”
A tall strip with each step labeled.
“A 4:1 banner of a mountain range at dawn for a newsletter header”
A wide strip sized for the header.
“Change the label on this jar (attached) to read Wild Honey, keep the lighting”
A new label, the rest of the photo untouched.
“Place this lamp (attached) in this living room (attached), evening light”
One composition from both references.
“A spec card for a pair of headphones with small feature callouts, 2K”
Fine print that holds up at full size.
Qwen Image 3 is one node on a canvas
Chain it with language, image, and video models, then publish the whole pipeline as an API. See how a full AI video pipeline fits together.
- Step 01
Draft the brief
A language model node expands a one-line idea into a detailed, on-brand prompt before it ever reaches the model.
- Step 02
Generate with Qwen Image 3
Run variants in parallel, compare against other models on the same prompt, and keep every result in run history.
- Step 03
Publish it as an API
The pipeline becomes a versioned endpoint. Your product calls it with the typed /invoke API or any OpenAI SDK.
Qwen Image 3 as an API, priced per image
The Qwen Image 3 API
Build a pipeline with Qwen Image 3 on a node and publish it. It becomes a versioned HTTP endpoint: the typed /invoke route enqueues a run and returns a run id to poll, and an OpenAI-compatible route works with SDKs you already use. Agents connect over MCP the same way. Your prompt engineering ships inside the pipeline, so every caller gets your quality, and swapping the model later does not change the API your product calls.
Qwen Image 3 pricing
Prepaid credits, no subscription. Each image is charged at Qwen Image 3's own rate from your balance, shown before you run, and only successful generations are charged, so a failed run costs nothing. Credit packs start at $5 and never expire, which suits spiky production schedules. Run history records the cost of every generation, per node, so spend is auditable rather than a surprise.
Qwen Image 3, answered
What is Qwen Image 3 best at?
Precision. Qwen built it to render text and details as small as 10 pixels, which makes it a strong pick for menus, labels, packaging, infographics, and any image with small print. It generates and edits in the same model.
What is the difference between Qwen Image 3 and Qwen Image 3 Pro?
Pro is the premium tier of the same model, with the same settings: text and image input, 1K or 2K output, thirteen aspect ratios, and up to 4 references. It is billed at a higher rate per image. Run one prompt through both on two nodes and let the output decide.
Which aspect ratios does Qwen Image 3 support?
Thirteen: 1:1, 4:5, 5:4, 3:4, 4:3, 2:3, 3:2, 9:16, 16:9, 1:2, 2:1, 1:4, and 4:1. Pick one on the node, along with 1K or 2K output.
How much does Qwen Image 3 cost on Treza?
Treza uses prepaid credits with no subscription. Each image is charged at Qwen Image 3's rate from your balance, and only successful generations are charged. Credit packs start at $5 and never expire.
Can I call Qwen Image 3 through an API?
Yes. Build a pipeline that uses Qwen Image 3 and publish it as a versioned HTTP endpoint. Call the typed /invoke API from your product, or point any OpenAI SDK at the OpenAI-compatible endpoint. If a better model ships later, swap it on the node without changing the API your product calls.
Your first video is one prompt away.
Generate your first video, image, or draft today. Prepaid credits, no subscription.