A 3-minute lip-sync reel on H3 Max: 13 chunks, $11.25 to $36.00

A 180 second track needs 13 MiniMax H3 Max lip-sync jobs because each accepts only 5 to 14.8 seconds of audio. Billed per whole audio second, the track costs 180 x $0.0625 = $11.25 at 480p, 180 x $0.10 = $18.00 at 768p and 180 x $0.20 = $36.00 at 1080p.

Why 13 jobs

The Models overview states the audio window as 5 to 14.8 seconds and the price as list x 1.25. Twelve chunks of 14 seconds cover 168 seconds, and a thirteenth chunk of 12 seconds finishes the track. Cutting at 14 whole seconds keeps every chunk inside the window and avoids partial-second rounding, since pricing is per ceil(audio second).

Cost by resolution

Resolution decides the rate, not the job count.

Planning the batch

Thirteen jobs fit inside a Pro workspace's accepted capacity of 24, but only four run at a time, so expect about four waves of processing. A Free workspace accepts 6 jobs, so a 13-job reel has to be submitted in at least three rounds.

Leave a split point at a natural pause in the speech so the cuts are not audible. A final chunk shorter than 5 seconds falls outside the audio window, so rebalance the last two chunks if the remainder is short. For example, a 171 second track splits into 12 chunks of 14 seconds plus a 3 second tail; make the last two chunks 8.5 and 8.5 seconds rather than 14 and 3.

Cost per chunk and per minute

Each 14 second chunk costs 14 x $0.0625 = $0.875 at 480p, 14 x $0.10 = $1.40 at 768p and 14 x $0.20 = $2.80 at 1080p. The 12 second tail is $0.75, $1.20 and $2.40. A full minute of audio is $3.75, $6.00 or $12.00 depending on the resolution.

These per-minute figures are handy for planning longer material. A 10 minute talking-head track at 768p is $60.00, while the same track at 480p is $37.50. Since each job is capped at 14.8 seconds, a 10 minute track is 43 jobs: 42 of 14 seconds and one of 12.

Lip-sync takes a Sume-hosted audio file, per the catalog description, so upload or generate the audio through Sume first and slice it into the chunks before submitting.

• Audio window: 5 to 14.8 seconds per job.
• Billing: per whole audio second, rounded up.
• Rates: 480p $0.0625, 768p $0.10, 1080p $0.20.