Luma Ray3.2 5 s at 1080p is $1.20: the same clip priced on Sume models
Luma lists Ray3.2 at $1.20 for 5 s at 1080p and $0.30 at 720p. Sume lists no Luma model; a 5 s 1080p clip is $0.70 to $7.11 on Sume models.

Luma's Ray 3.2 price table lists a 5 second standard-range clip at $0.30 for 720p and $1.20 for 1080p. Sume's video docs list no Luma model, so the nearest way to get a 5 second clip is one of Sume's own ids: at 1080p, Kling 3 costs $0.70 to $1.05, Gemini Omni Flash 1.1 $0.94, MiniMax H3 Max $1.00, Wan 3.0 $1.25, Seedance 2.0 $4.25 and Seedance 2.5 $7.11. At 720p Luma is cheaper than all of them.
Luma's published rates
Luma's pricing page lists these standard dynamic range rates for clips from text or image. A 10 second clip costs three times a 5 second one at the same resolution, so two 5 second clips are cheaper than one 10 second clip.
| Resolution | 5 seconds | 10 seconds |
|---|---|---|
| 360p (draft) | $0.06 | $0.18 |
| 540p | $0.15 | $0.45 |
| 720p | $0.30 | $0.90 |
| 1080p | $1.20 | $3.60 |
The same 5 seconds on Sume
Sume bills the provider list x 1.25 for each output second (Seedance by video tokens, with the same multiplier). Five seconds at 1080p is 5 x the per-second rate below.
| Sume id | Resolution | Sume per second | 5 s at 1080p | 5 s at 720p |
|---|---|---|---|---|
| gemini-omni-flash-1.1 | 720p / 1080p | $0.125 / $0.1875 | $0.9375 | $0.625 |
| minimax-h3-max | 768p / 1080p | $0.10 / $0.20 | $1.00 | $0.50 (768p) |
| kling-3 | 720p / 1080p | $0.14 to $0.21, flat | $0.70 to $1.05 | $0.70 to $1.05 |
| wan-3.0 | 720p / 1080p | $0.125 / $0.25 | $1.25 | $0.625 |
| seedance-2 | 720p / 1080p | $0.378 / $0.8505 | $4.2525 | $1.89 |
| seedance-2.5 | 720p / 1080p | $0.5778 / $1.42155 | $7.11 | $2.889 |
What the numbers do and do not say
Luma at 720p ($0.30) is less than half of any Sume row at 720p. At 1080p the gap is small: $1.20 against $0.94 to $1.25 for the cheaper Sume ids. Price is one input. Luma's page lists clips up to 20 seconds and up to 16 keyframes per clip, and its video-to-video product works from a source clip. Sume's ids take a first and last frame, not 16 keyframes.
Pick by the limit that binds
If you only need a 5 second 1080p hook, the price spread between Luma, Omni, H3 Max and Wan is about 30 cents, and the choice should come from which clip looks right. If you need clips of 21 to 30 seconds, Wan 3.0 and Seedance 2.5 take them in one job on Sume, while Luma's maximum is 20. For a table of what each Sume id accepts, see the reference limits post.
What to run first
For a 5 second 1080p hook the spread between the cheaper Sume ids is small, so run one prompt through Kling 3, Gemini Omni Flash 1.1 and Wan 3.0 and keep the best. The three cost $1.05 + $0.9375 + $1.25 = $3.2375 at 1080p, or $0.35 less if you leave Kling's audio off.
If a 720p draft is enough to judge the idea, Wan and Omni cost $0.625 each and Kling $0.70 to $1.05, so the three cost $1.95 to $2.30. Luma's $0.30 at 720p is cheaper still, but Sume does not list Luma, so you would buy that clip from Luma directly.
Sources
Related posts
More in Comparisons
- MAI-Transcribe-2-Streaming is a preview with no SLA: add a fallback
Microsoft marks MAI-Transcribe-2-Streaming public preview with no SLA. Record call audio and keep Sume STT as a recorded fallback at about one cent per minute.
- MAI-Transcribe-2-Streaming vs Sume STT: 100 hours, $54 vs $60
A tracker lists MAI-Transcribe-2-Streaming at $0.54 per hour (introductory through 2026) and Sume STT is $0.01 per audio minute, or $0.60 per hour.
- MAI-Voice-2.1 and MAI-Transcribe-2: what Sume lists for voice and STT
Microsoft AI's news page names MAI-Voice-2.1 and MAI-Transcribe-2. Sume lists neither; it lists Sonic TTS (38 micros a character) and STT at $0.01 a minute.
- MAI-Voice-2.1 or Sume TTS for a presenter ad read: 3 prices
For a 700-character ad read, MAI-Voice-2.1 is about 1.5 cents, Flash 1 cent and Sume TTS 4 cents. Fabric needs the audio as a public URL.
Written by Sume