Sume's image router lists 17 model ids on Oct 8, 2026; no Hy Image
The public Sume catalog on 2026-10-08 lists 17 model ids under Image Router. None is Tencent's Hy Image 3.5 Preview. See the list and the check.
The short answer
Sume does not list Tencent's Hy Image 3.5 Preview. The public catalog at api.sume.com/v1/catalog, read on 2026-10-08, shows 17 model ids under the Image Router entry, and none of them is a Tencent or Hy model. OpenRouter, by contrast, shows the model with the id dated 20260922.
The two sources give different dates for the same preview, so cite the one you actually read.
The arithmetic
The 17 ids in the catalog entry are gpt-image-2.5, gpt-image-2.5-sunburst, gpt-image-2, nano-banana-2, nano-banana-pro, seedream-5-lite, seedream-4-5-edit, grok-image, seedream-v4, qwen-image, imagen-4-fast, imagen-4-ultra, flux-2-pro, flux-2-flex, ideogram-v3, recraft-v4 and qwen-image-max.
The table sets the Hy facts that OpenRouter publishes next to what the Sume catalog and docs show. Sume's docs cover its own list, so only the left column is a Sume claim.
| Item | Hy Image 3.5 Preview | Sume |
|---|---|---|
| Listed on Sume catalog | No | 17 ids under Image Router, none Tencent |
| References per request (OpenRouter page) | 20 | GPT Image 2.5 on Sume takes up to 16 |
| Images per request (OpenRouter page) | n limited to 1 | Sume n range is per model, up to 10 |
| Resolution labels (OpenRouter page) | 1K, 1.5K, 2K, 4K | Sume tiers are 512, 1K, 2K, 4K |
| Id on OpenRouter | includes 20260922 | Not applicable on Sume |
How to run it
Check it yourself with one request: fetch the catalog URL, which needs no key, and search the Image Router model_ids for any Tencent or Hy id. If you do not find one, Sume does not list it.
If you need an image model that Sume lists, the closest feature match by reference count is GPT Image 2.5 at 16, and the closest by price is GPT Image 2.5 low at $0.02475 per image.
Retries and what to watch
The check in this post is cheap to repeat, so repeat it whenever a launch is announced. The catalog is public and needs no key, and the id list is in the Image Router entry.
Three outcomes are possible when you look.
- The id is listed: read its endpoint record for prices and parameters before you send a call.
- The id is not listed: Sume does not offer the model, so pick a listed row or wait for a listing.
- A similar id is listed: compare the supported_parameters, since a different model with a similar name can take different values.
What the docs say to check
Read GET /v1/images/models before you pin a tier or ratio. A model accepts only the values its descriptors list, and the descriptors are the definitive source for what a call may send.
Do not assume that a model on an aggregator is on every aggregator. Preview models in particular can change names or leave, so confirm the catalog on the day you plan a job.
Every price in this post is a snapshot dated 2026-10-08. The Sume figures come from the public catalog and the Image API docs, the Google figures come from its Gemini API pricing page, and no figure here is a benchmark of quality or speed. Prices and model lists change, so confirm them on the day you spend money.
Sources
Related posts
More in Models
- sume/music-auto or pinned lyria-3.5: what changes in the job
Music Router takes sume/music-auto (Lyria 3.5 today), lyria-3.5 or lyria-3-pro. What job.model and routed_model echo back, and why the price stays the same.
- Does the Sume spend cap cover the agent's own model tokens?
No: Sume's cap excludes the agent's LLM turn; debited_usd_micros includes it. At list rates that turn is about $0.004 on Haiku 5.5 and $0.40 on Astra.
- Swap four people in one video: H3 Max Recast bills per second
h3-max-recast takes one source video and 1 to 4 person photos, 5 to 30 seconds. Price is $0.375 or $0.5625 a second whether you swap one person or four.
- Text-in-image bake-off: 20 prompts on 4 Sume models costs $7.00
Test text rendering in AI images for $7.00: 20 prompts on Ideogram 4.5, GPT Image 2.5, Qwen Image Max and Nano Banana 2.1, with the cost of each on Sume.
Written by Sume