Sume POST /v1/images (sume/auto) vs Image 1.0: which for new code
Sume's docs say Image 1.0 is retiring and is an alias for Image Router Auto. New code should use POST /v1/images. The field differences.
Use POST /v1/images for new code. Sume's docs say Image 1.0 is retiring soon and is a compatibility alias for Image Router Auto: its legacy URLs keep accepting the old request shape but run the same Auto model selection and return job.model sume/auto. You can also pick a model on /v1/images by sending a catalog model id instead of sume/auto.
Field differences
Names below are taken from the Image 1.0 docs and the Image API docs.
| Need | POST /v1/images | POST /v1/image-1.0/generate |
|---|---|---|
| Choose a model | Catalog id, or model: sume/auto | Always Auto |
| Reference images | input_references (0 to 10 on the example model) | image_urls, 1 to 10 URLs |
| Image count | n, 1 to 4 in the example | num_images, 1 to 4 |
| Masked edit | mask_url on models that support it | mask_image_url with image_urls |
| Quality | Model-specific; gpt-image-2.5 default is high | low (default), medium, high |
| Discovery | GET /v1/images/models | Not described on that page |
The default-quality trap
On Image 1.0 the quality field defaults to low. On the Image API with gpt-image-2.5 the docs say that if you omit quality the default is high. The price differs by tier: gpt-image-2.5 is $0.02475 at low 1K and $0.2225 at high 4K in the Sume catalog, so a migration that drops quality can raise cost. Send quality explicitly.
Image 1.0 uses only public HTTPS image URLs and rejects localhost and private-network URLs before submission.
Migration checklist
Rename image_urls to input_references, num_images to n and mask_image_url to mask_url where the model supports them, then choose between sume/auto and a pinned catalog id. Verify each model's parameters with GET /v1/images/models/{model_id}/endpoints, as the Image API docs describe. The old URLs are described in the Image 1.0 docs.
Sources
More in Comparisons
- Sume TTS 1.0 vs TTS Router: same $0.0475 per 1,000 characters
Sume's catalog prices TTS 1.0 and TTS Router identically: $0.0475 per 1,000 characters, 20,000-character maximum. They differ in how you pick the model.
- Sume TTS is a job, not a stream: cost and fit for a voice agent reply
A 45-second spoken reply is about 675 characters: $0.0320625 on Sume TTS, $0.010125 on MAI-Voice-2.1-Flash. Sume returns a finished file, not a live stream.
- Switching Fabric to H3 Max Lip Sync: 6 s, 12 s and 20 s clips compared
Fabric and H3 Max Lip Sync share a body on Sume, so switching is a URL change. At 6 s H3 768p is $0.60 vs Fabric 720p $1.125; at 20 s H3 Max is not eligible.
- Synthesia Basic free 10 minutes vs 10 minutes of Sume avatar video
Synthesia Basic is $0 with 10 minutes and 9 avatars; the same 10 minutes on Sume standard is $110.40, so the free plan wins until you outgrow it.
Written by Sume