Hy Image 3.5 Preview. Tencent's image model with up to 20 references.
Tencent's unified generation and editing model. Make an image from a prompt, or edit and compose from up to 20 reference images, in thirteen aspect ratios and four resolution settings.
- Text and image input
- 1K, 1.5K, 2K or 4K setting
- 13 aspect ratios, 1:4 to 4:1
- Up to 20 reference images
Hy Image 3.5 Preview is Tencent's image model on OpenRouter, listed on 2026-10-05. One model handles text-to-image and image-to-image: give it a prompt, or a prompt plus up to 20 reference images, and pick a resolution from 1K to 4K. In our tests a square came back at 1024, 2048 or 4096 pixels to match the setting, and 1536 pixels with no setting at all.
On Treza it sits in the Image Generation node's picker and in chat. Renders took 17 to 23 seconds at 1K and 2K, 44 seconds at 4K, and 52 seconds for an edit with one reference. The reference edit kept the product and moved it into the new scene. It bills from prepaid credits, and only successful runs are charged.
Up to 20 references
Carry a product, a character and a setting into one image. In our tests a reference image added nothing to the price.
Four resolution settings
1K, 1.5K, 2K or 4K. A square rendered at 1024, 2048 and 4096 pixels for the settings we tried, and 1536 with none set.
One price up to 2K
In our tests a 1K, 1.5K and 2K image each billed the same, and 4K cost about a third more.
Thirteen aspect ratios
From 1:4 and 9:16 portraits to 16:9 and 4:1 banners, with a seed for repeatable results.
What to make with Hy Image 3.5 Preview
Prompts to try, and what to expect back. Each runs on an Image Generation node with Hy Image 3.5 Preview picked in the model menu.
“A ceramic cup of coffee on a sunlit wooden cafe table, steam rising, shallow depth of field”
A warm product-style still.
“The same folded shirt (attached) on a white marble counter in a bright studio”
Your product moved into a new set.
“This sofa (attached) in this living room (attached), afternoon light”
Two references composed into one scene.
“A 4:1 banner of a coastal road at golden hour for a website header”
A wide still at the ratio you asked for.
“Keep this character (attached) and show her at a market, a station and a rooftop”
One character across a series.
“A 4K print-ready still of a lighthouse on a cliff in a storm”
A 4096-pixel square for print or crop.
Hy Image 3.5 Preview 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 Hy Image 3.5 Preview
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.
Hy Image 3.5 Preview as an API, priced per image
The Hy Image 3.5 Preview API
Build a pipeline with Hy Image 3.5 Preview 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.
Hy Image 3.5 Preview pricing
Prepaid credits, no subscription. Each image is charged at Hy Image 3.5 Preview'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.
Hy Image 3.5 Preview, answered
What does Hy Image 3.5 Preview cost per image?
OpenRouter lists it per output image token. In our tests a 1K, 1.5K and 2K image each billed 15,000 tokens, and a 4K image 20,000, with a reference image adding nothing. Treza quotes those measured amounts, plus its credit markup, before a run.
How many reference images can I attach?
Up to 20. Wire them into the Image Generation node's reference input, or attach them in chat, and describe how they should come together.
Can I use it in chat?
Yes. Ask for Hy Image 3.5 by name in chat, attach reference images if you have them, and the image comes back in the conversation.
How much does Hy Image 3.5 Preview cost on Treza?
Treza uses prepaid credits with no subscription. Each image is charged at Hy Image 3.5 Preview's rate from your balance, and only successful generations are charged. Credit packs start at $5 and never expire.
Can I call Hy Image 3.5 Preview through an API?
Yes. Build a pipeline that uses Hy Image 3.5 Preview 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.