Upload 30 reference images to a Sume Format run (images only)
A Format run takes up to 30 images in attachments[], type input_image, by public HTTPS URL or asset_id. Video and audio attachments are not supported today.

A Sume Format run accepts up to 30 images as attachments[] on the create body, each with type: "input_image". That is the only attachment type today, so videos and audio files are not attachments, unlike Opus Clip's Motion Studio, which takes up to 10 videos and 10 audio files alongside its images. Video and audio references go by URL in input instead.
Limits are from the Formats docs, read 2026-10-01; the Opus Clip counts are from its Sept 18 changelog entry.
What does an attachment look like?
Each entry is an object with a type and exactly one image source. Fields as the docs list them:
| Field | Required | Notes |
|---|---|---|
type | Yes | input_image, the only type today |
image_url | One of | Public HTTPS URL; Sume fetches it when the run is created, so it must be reachable without auth |
asset_id | One of | An image uploaded through the Assets API, ready, in the same workspace |
How do I send a run with attachments?
Put them beside instruction and input, for example { "instruction": "Make a product hero from the attached photo.", "attachments": [{ "type": "input_image", "asset_id": "asset_..." }] }. filename is optional on an entry. Use asset_id for private images, since an image_url is fetched without credentials.
What if my references are videos or audio?
They are not attachable. Extract stills from a clip with video_inspect or the frames endpoint and attach those as images, or use a media step in the Format itself. For plain image generation with references see image generation API reference images.
Does the agent treat attachments as instructions?
The docs describe attachments as images the agent can look at. Put your requirements in instruction, which comes after the Format's own text, and keep the pictures as reference only.
Sources
Related posts
More in Developers
- Gemini 2.5 not available in a new project: what to do
Google's September 18 changelog limits Gemini 2.5 models to users who already used them. New projects are pointed to 3.5 Flash-Lite or 3.8 Flash.
- gemini-3.1-flash-tts-preview replacement: 3.8 Flash-Lite TTS
Google names gemini-3.8-flash-lite-tts as the replacement for the 3.1 preview TTS model. What that changes for you, and what stays fixed on Sume TTS.
- Gemini TTS API polling: the Sume audio job pattern
Gemini 3.8 Flash TTS went GA on Sep 22. On Sume, speech is a job: submit async, store the job id, poll the status URL, and never resubmit while it runs.
- Gemini 3.8 TTS returns WAV by default; Sume TTS returns MP3
Gemini 3.8 TTS unary requests return WAV with a RIFF header. Sume TTS 1.0 defaults to mp3 at 44100 Hz, so ask for wav when your pipeline needs it.
Written by Sume