SeedVR2 10x video upscale vs Sume's upscale tool
SeedVR2 is listed with 1x to 10x video upscaling. Sume exposes a video_upscale_create tool, but its docs name no scale factors, so check the live schema first.

fuser.studio's upscaler roundup lists SeedVR2 for video upscaling at 1x to 10x scaling and places it in the generative class. Sume exposes a video_upscale_create tool on its hosted MCP, but the public docs do not state its scale factors, so read the tool schema before you plan a 10x job.
What generative upscaling means
Generative upscalers synthesize detail rather than only sharpening what is there. That can recover texture on soft footage, and it can also invent detail that was never in the source, such as skin texture or text. The more aggressive the factor, the more the output depends on the model's guess.
What Sume documents
video_upscale_create appears in the hosted MCP tool list next to image_upscale_create and rmbg_create. The video generation docs also say 2K and 4K upscales are priced when requested on minimax-h3. Nothing in the docs gives a maximum factor or a list of upscale models for the tool, so do not assume parity with SeedVR2.
Check the schema before you commit
Ask the server instead of guessing. tools_schema returns the input schema for a named tool, and tools_list shows the subset visible to your session:
- Call
tools_listand confirmvideo_upscale_createis visible to your session - Call
tools_schemafor it and read the accepted fields - Run a dry run (
dry_run=true) to preflight cost before a paid submit - Keep an
idempotency_keyon every paid call
Generate at the size you need
If you need an upscale beyond what the tool offers, a pragmatic route is to generate at a higher resolution in the first place. Several catalog models list 1080p, and Gemini Omni Flash 1.1 lists 4K, which avoids a second pass altogether.
Sources
Related posts
More in Media tools
- Split one recording into many audio clips: detach once, then split
Twenty ranges from one talking-head video need one audio detach and one split job, not twenty cuts. Wav or mp3, overlap rules, and what each clip returns.
- Stability AI's Series B and label backers: what it means for audio
Stability released Stable Audio 3.0 on 5/20/26 and raised a Series B on 8/25/26 with EA, Sony, UMG and WMG. A dated timeline and what to verify for video work.
- Stabilize shaky video by API: Sume has no deshake, so do this
Sume's video filter refuses deshake and vidstab as unknown filters. Prove it with the free check, stabilize upstream, then use Sume for trim, crop and captions.
- Temu detail video up to 180 s: stitch 30 s clips with Timeline
Temu detail video allows up to 180 s and 300 MB at 720p or higher. Sume clips top out at 30 s, so six clips stitched with Timeline cover the full length.
Written by Sume