Models
The models behind Sume's video, image, speech, and music endpoints: ids, clip lengths, resolutions, aspect ratios, audio, references, and migration.
Start with What AI models does Sume support? Every model id, by API surface
MiniMax H3 has no 1080p on Sume: use H3 Max at $0.20 a second
minimax-h3 on Sume tops out at native 768p (2K and 4K are priced upscales). minimax-h3-max has 1080p at $0.20 a second, refined up from 768p.
MiniMax H3 audio references: 2 to 15 s each, 15 s combined
On Sume, MiniMax H3 audio references must be 2 to 15 seconds each, 15 seconds combined, at most 3 files, and never the only reference. Here is how to prep them.
Which Sume video ids turn audio off, and which add sound
Gemini Omni 1.1, MiniMax H3 and H3 Max always produce audio on Sume; generate_audio false is rejected. Kling 3 prices sound on and off, Wan reads a flag.
Omni Flash edit on Sume: what editing a 6 s and a 30 s clip costs
Editing a clip with Omni Flash on Sume bills per output second at the resolution you pick: $0.75 for 6 s at 720p, up to $3.75 for a 30 s hint. How it is priced.
Sonic 3.6, 3.5 and 3 cost the same on Sume TTS Router: pick by ear
All five Sume TTS Router ids share one per-character price, so model choice is a listening test, not a budget one. Here is how to run it for a few cents.
TikTok's three ad ratios and Kling 3's three: 9:16, 16:9 and 1:1
TikTok's non-Spark ad specs list 9:16, 16:9 and 1:1. Kling 3 on Sume lists exactly those three, at 720p or 1080p and 4 to 15 seconds. Here is what that fits.
Wan 3.0 reference videos: 5 clips, 15 s total, 16 fps floor
Wan 3.0 on Sume takes up to 5 reference videos, 15 seconds combined, each at 16 fps or higher. Check clip count, length and frame rate before you submit.
GPT Image 2.5: image_size wins over aspect_ratio, plus the pixel rules
Send aspect_ratio 1:1 and image_size 1280x720 to GPT Image 2.5 on Sume and the pixels win. The multiples-of-16, 3:1 and pixel-count rules, with a validator.
Which Sume image models reject input_references: five text-only rows
Five of 19 Sume image models are text-to-image only and return 400 unsupported_parameter if you send input_references. The other 14 take 5, 10 or 16 references.
Claymation look test: four Sume video models at low resolution
Test a claymation prompt on Omni Flash 1.1, Wan 3.0, H3 and Seedance 2.0 Mini for under $1.20 in drafts, then price the finals at 1080p before committing.
One narrator across a series: Sume TTS 1.0 is Sonic 3.6, pin older ids
Sume TTS 1.0 always runs on sonic-3.6. To keep a series that began on an older Sonic, use the TTS Router and pin sonic-3.5 or sonic-3 with the same voice.
Neon cyberpunk street at 1080p for 10 seconds: five Sume models
A 10-second 1080p cyberpunk street clip costs $1.40 to $2.50 on Kling 3, Omni Flash 1.1, H3 Max and Wan 3.0 on Sume; Seedance 2.5 is $14.22.
Is MiniMax H3 Max 1080p native? Sume's catalog and price
H3 Max renders natively at 480p and 768p; its 1080p is a latent refinement from 768p. On Sume 1080p is $0.20 a second, double the 768p rate.
Gemini Omni Flash 1.1 3-second clip: 360p to 4K price
A 3-second Gemini Omni Flash 1.1 clip on Sume costs $0.1125 at 360p, $0.3750 at 720p, $0.5625 at 1080p and $1.1250 at 4K.
Gemini Omni Flash 1.1 reference-to-video: IMAGE_REF tags
Use up to 10 reference images and 3 short reference videos in one Gemini Omni Flash 1.1 request on Sume, and point at them with <IMAGE_REF_0> and <VIDEO_REF_0>.
Gemini Omni Flash 1.1 6-second clip: 360p to 4K price
A 6-second Gemini Omni Flash 1.1 clip on Sume costs $0.2250 at 360p, $0.75 at 720p, $1.1250 at 1080p and $2.25 at 4K.
Gemini Omni Flash 1.1 9-second clip: 360p to 4K price
A 9-second Gemini Omni Flash 1.1 clip on Sume costs $0.3375 at 360p, $1.1250 at 720p, $1.6875 at 1080p and $3.3750 at 4K.
Gemini Omni Flash 1.1 5-second clip: 360p to 4K price
A 5-second Gemini Omni Flash 1.1 clip on Sume costs $0.1875 at 360p, $0.6250 at 720p, $0.9375 at 1080p and $1.8750 at 4K.
Edit an existing video with Gemini Omni Flash 1.1 over an API
Send video_url and an instruction to gemini-omni-flash-1.1 on Sume to edit a clip: no duration or aspect ratio, resolution optional, native audio on.
Gemini Omni Flash 1.1 7-second clip: 360p to 4K price
A 7-second Gemini Omni Flash 1.1 clip on Sume costs $0.2625 at 360p, $0.8750 at 720p, $1.3125 at 1080p and $2.6250 at 4K.
Gemini Omni Flash 1.1 8-second clip: 360p to 4K price
A 8-second Gemini Omni Flash 1.1 clip on Sume costs $0.30 at 360p, $1.00 at 720p, $1.50 at 1080p and $3.00 at 4K.
Gemini Omni Flash 1.1 4-second clip: 360p to 4K price
A 4-second Gemini Omni Flash 1.1 clip on Sume costs $0.15 at 360p, $0.50 at 720p, $0.75 at 1080p and $1.50 at 4K.
H3 Max Recast on a 25-second video: person swap price
Swapping people in a 25-second source video with H3 Max Recast on Sume costs $9.3750 at 768p and $14.0625 at 1080p.
MiniMax H3 vs H3 Max for a 10-second clip: 480p and 768p
A 10-second clip on Sume: H3 $0.6250 at 480p and $0.75 at 768p; H3 Max $0.6250 and $1.00.
H3 Max Recast on a 12-second video: person swap price
Swapping people in a 12-second source video with H3 Max Recast on Sume costs $4.50 at 768p and $6.75 at 1080p.
5 Kling 3 clips of 10 seconds with audio: $10.50
5 ten-second Kling 3 clips with audio on Sume cost $10.50 ($2.10 each). Without audio each is $1.40.
H3 Max Recast on a 5-second video: person swap price
Swapping people in a 5-second source video with H3 Max Recast on Sume costs $1.8750 at 768p and $2.8125 at 1080p.
50 Kling 3 clips of 10 seconds with audio: $105.00
50 ten-second Kling 3 clips with audio on Sume cost $105.00 ($2.10 each). Without audio each is $1.40.
H3 Max Recast on a 15-second video: person swap price
Swapping people in a 15-second source video with H3 Max Recast on Sume costs $5.6250 at 768p and $8.4375 at 1080p.
MiniMax H3 vs H3 Max for a 7-second clip: 480p and 768p
A 7-second clip on Sume: H3 $0.4375 at 480p and $0.5250 at 768p; H3 Max $0.4375 and $0.70.
MiniMax H3 vs H3 Max for a 15-second clip: 480p and 768p
A 15-second clip on Sume: H3 $0.9375 at 480p and $1.1250 at 768p; H3 Max $0.9375 and $1.50.
H3 Max Recast on a 20-second video: person swap price
Swapping people in a 20-second source video with H3 Max Recast on Sume costs $7.50 at 768p and $11.25 at 1080p.
10 Gemini Omni Flash 1.1 clips of 8 seconds at 720p: $10.00
10 eight-second Gemini Omni Flash 1.1 clips on Sume at 720p, $1.00 each, $10.00 total, with synced audio included. 1080p and 4K compared.
25 Kling 3 clips of 10 seconds with audio: $52.50
25 ten-second Kling 3 clips with audio on Sume cost $52.50 ($2.10 each). Without audio each is $1.40.
MiniMax H3 vs H3 Max for a 8-second clip: 480p and 768p
A 8-second clip on Sume: H3 $0.50 at 480p and $0.60 at 768p; H3 Max $0.50 and $0.80.
H3 Max Recast on a 30-second video: person swap price
Swapping people in a 30-second source video with H3 Max Recast on Sume costs $11.25 at 768p and $16.8750 at 1080p.
25 Gemini Omni Flash 1.1 clips of 8 seconds at 720p: $25.00
25 eight-second Gemini Omni Flash 1.1 clips on Sume at 720p, $1.00 each, $25.00 total, with synced audio included. 1080p and 4K compared.
Gemini Omni Flash 1.1 generate_audio false: why the API refuses
gemini-omni-flash-1.1 on Sume always makes native synced audio, so generate_audio: false is rejected. Replace the sound in a Timeline instead.
5 Gemini Omni Flash 1.1 clips of 8 seconds at 720p: $5.00
5 eight-second Gemini Omni Flash 1.1 clips on Sume at 720p, $1.00 each, $5.00 total, with synced audio included. 1080p and 4K compared.
50 Gemini Omni Flash 1.1 clips of 8 seconds at 720p: $50.00
50 eight-second Gemini Omni Flash 1.1 clips on Sume at 720p, $1.00 each, $50.00 total, with synced audio included. 1080p and 4K compared.
MiniMax H3 vs H3 Max for a 6-second clip: 480p and 768p
A 6-second clip on Sume: H3 $0.3750 at 480p and $0.45 at 768p; H3 Max $0.3750 and $0.60.
MiniMax H3 vs H3 Max for a 5-second clip: 480p and 768p
A 5-second clip on Sume: H3 $0.3125 at 480p and $0.3750 at 768p; H3 Max $0.3125 and $0.50.
MiniMax H3 vs H3 Max for a 9-second clip: 480p and 768p
A 9-second clip on Sume: H3 $0.5625 at 480p and $0.6750 at 768p; H3 Max $0.5625 and $0.90.
H3 Max Recast on a 8-second video: person swap price
Swapping people in a 8-second source video with H3 Max Recast on Sume costs $3.00 at 768p and $4.50 at 1080p.
MiniMax H3 vs H3 Max for a 12-second clip: 480p and 768p
A 12-second clip on Sume: H3 $0.75 at 480p and $0.90 at 768p; H3 Max $0.75 and $1.20.
10 Kling 3 clips of 10 seconds with audio: $21.00
10 ten-second Kling 3 clips with audio on Sume cost $21.00 ($2.10 each). Without audio each is $1.40.
1,000 SKU catalog of 6 s Grok Imagine loops: $75 plus image prep
Grok Imagine Video on Sume bills $0.0125 a second: 1,000 six-second loops are $75.00, with no audio or aspect_ratio. Add 9:16 prep photos and submit in waves.
Lyria 3.5 has no edit pass: iterate a music bed for $0.125 a take
Google says Lyria 3.5 is single-turn. Iterate a bed on Sume by changing one prompt axis per take; five takes cost $0.625 and a retry needs a key.