Firefly Upscale API vs Sume's image_upscale_create tool

Firefly's Upscale API enhances image resolution and quality, with 2x, 4x or 6x scale factors in its user-assets use case. On Sume, image upscaling is the image_upscale_create tool in the hosted MCP server: a paid write tool that needs an idempotency_key and returns a job you read with the standard status and result calls.

Adobe facts are from its API overview; Sume facts from MCP tools and gates and Jobs and results, read 2026-10-01.

What does the Firefly Upscale API cover?

Adobe lists two use cases. User Assets Upsampler increases the resolution of your own photos and artwork while following the source closely, at 2x, 4x or 6x. Generated Image Upsampler keeps crispness and coherent fine detail, such as fabric or grass textures, in AI-generated images.

How does upscaling work on Sume?

The MCP tools page lists image_upscale_create among the paid tools that take an idempotency_key. Like other Sume jobs, you then poll status and fetch the result. The Image API docs name the same pattern: poll GET /v1/jobs/{id}/status and fetch GET /v1/jobs/{id}/result.

Should I wait inside one call?

No. The jobs docs say "30 seconds is a wait budget, not a job duration". If a call returns before the job is terminal, keep the job id and read it again rather than resubmitting a paid request. See AI image upscale API for the broader picture.

Which one fits my workflow?

If you already run inside Adobe's service and need a specific scale factor, the Firefly page is the reference. If your pipeline already talks to Sume through MCP or REST jobs, the upscale tool keeps the same job model as generation. Check the tool's input schema in your MCP client for the parameters it accepts.