11-second H3 Max clip: $0.69 at 480p, $1.10 at 768p, $2.20 at 1080p

An 11-second MiniMax H3 Max clip costs $0.69 at 480p, $1.10 at 768p and $2.20 at 1080p on Sume, with native stereo audio included. The plain MiniMax H3 row is $0.69 at 480p and $0.83 at 768p.

Price table at 11 seconds

Eleven seconds is a legal whole-second length for both rows, which accept 5 to 15 seconds. Billing is list times 1.25, rounded up to the cent.

Arithmetic for each row

H3 Max lists at $0.05 per second at 480p, $0.08 at 768p and $0.16 at 1080p. For 11 seconds that is $0.55, $0.88 and $1.76 list. After the 1.25 margin they are $0.6875 (billed $0.69), $1.10 and $2.20. MiniMax H3 lists at $0.05 and $0.06, giving $0.55 and $0.66 list, $0.6875 and $0.825 after margin, billed $0.69 and $0.83.

Why 11 seconds

Eleven seconds fits a single spoken line and a beat after it, so the clip does not end mid-sentence. The audio is native stereo and is always generated; there is no generate_audio switch on either row, and sending generate_audio: false is refused.

Longer than 15 seconds is not available on these rows. For anything longer, make several clips and join them on a timeline.

• Pick 480p to test a prompt, 768p for most deliverables, 1080p for a final.
• 1080p on H3 Max is a latent refinement from native 768p, not a separate native render.
• Some 1080p image-to-video clips have a crop jump after two frames; check the opening frames before assembly.

Where 11 seconds sits on the ladder

Each extra second at 1080p costs $0.20 on H3 Max, so 11 seconds is $0.20 more than 10 seconds and $0.20 less than 12.

Examples of 11-second scripts

Eleven seconds is enough for a hook, one claim and a closing line. A spoken script of about 25 to 30 words fits at a normal pace, leaving a beat at the end. For a silent product beat, 11 seconds supports two camera moves of about five seconds each. If the clip is part of a longer piece, it matches a timeline slot of 11 seconds with no trimming. Drafting works like this: write the prompt, run it at 480p for $0.69, and look at the audio as well as the picture, since the sound is native and part of the result. Then re-run at 768p for $1.10 once the prompt is stable.

One more use for the rule: when you write a client quote, quote per clip, not per second, and say that each clip is rounded up to the cent. A quote of 50 clips at $0.53 each is $26.50, and it will match the invoice exactly, while a per-second estimate will be off by a few dollars.