CampaignReel

Ad placement specs against what generators output

Speech sets one length and a parameter sets the other

One page publishes a ten minute maximum on output driven by a supplied audio track. The other publishes an integer range of two to thirty seconds set by a request parameter. Both are lengths in seconds and neither is the same quantity as the other. As of 2026-09-22.

A bound on speech and a bound on coherenceWhere the picture follows an audio track, the ceiling describes file handling. Where it is generated from a prompt, it describes how long a scene holds together.Audio suppliedTen minutesLength follows the speechA handling boundParameter setTwo to thirty secondsLength follows the fieldA coherence boundSame unitSecondsOn both rowsDifferent quantitiesTwenty times apart, and not comparableWhy this column is never sorted on these sheets
Fig. 1 Sorting these into one order would put a file-handling bound above a generation bound and call it capability.
What decides how long the output runs. Recorded 2026-09-22.
What the pages stateHedraWan 3.0
What sets the lengthHow long the supplied audio runsA duration parameter
Published ceiling10 minutes30 seconds
What the ceiling describesA bound on file handlingHow long one scene holds together

Inclusion rule. Length figures each page states, with what each page says decides them. No ordering is implied between the two figures. Order. The mechanism first, then the figure it bounds.

1A ceiling on speech and a ceiling on coherence

Where the picture follows an audio track, the output runs as long as the track and the published figure is about what the service will handle. Where the picture is generated from a prompt, the figure is about how long a scene can be held together at all.

Sorting those into a single order would put a file-handling bound above a generation bound and call it capability. That is the category error this register is arranged to avoid, and it is the reason the mechanism sits in the table.

2For a sixty second slot both are one call, differently

One reaches a minute because a minute of speech is a small file. The other reaches it in two requests and one join, which is the fewest of any prompt-driven row here.

The choice between them is therefore not about the length at all. It is about whether the advertisement is a person speaking or a scene playing, and that is decided in the brief rather than in this column.

3The other columns tell a plan more than this one does

One page publishes seven shapes and three resolution rungs with no condition attached to any of them. The other publishes three tiers, six named shapes and an adaptive default that hands the shape decision to the model unless a request takes it back.

That adaptive default is the fact a campaign should read first on the second row, and neither the ten minute figure nor the thirty second one hints at it. The length column is the least informative thing these two rows share.

4The rest of the column behind these two

Duration across the registerHedra is marked; Wan 3.0 sits at an integer range to thirty seconds, set by a parameter in the same column. Recorded 2026-09-22.Duration across the registerDurationAmazon Nova Reel — Duration: 6 s increments, up to two minutesAmazon Nova ReelCogVideoX — Duration: 6 s, at 49 framesCogVideoXGoogle Veo — Duration: 4, 6 or 8 sGoogle VeoGrok Imagine — Duration: 1 to 15 sGrok ImagineHedra — Duration: 10 minutesHedraHeyGen — Duration: 30 minutes per sceneHeyGenKling AI — Duration: Not statedKling AILTX-Video — Duration: Frames, never secondsLTX-VideoMiniMax — Duration: 4 to 15 sMiniMaxPika — Duration: 5 s and 10 s, 20-25 s on PikaframesPikaPixVerse — Duration: 1 to 15 s on version 6PixVerseRunway — Duration: 5 to 15 s on Gen-4.5RunwaySceneMixer — Duration: Thirty seconds on the Wan tiers, fifteen on H3SceneMixerVidu — Duration: Not statedViduWan 3.0 — Duration: 2 to 30 sWan 3.0
Fig. 2 Hedra is marked; Wan 3.0 sits at an integer range to thirty seconds, set by a parameter in the same column. Recorded 2026-09-22.
Every generator in the register on the duration column, with Hedra marked. Recorded 2026-09-22.
GeneratorDuration
Amazon Nova Reel6 s increments, up to two minutes
CogVideoX6 s, at 49 frames
Google Veo4, 6 or 8 s
Grok Imagine1 to 15 s
Hedra10 minutes
HeyGen30 minutes per scene
Kling AINot stated
LTX-VideoFrames, never seconds
MiniMax4 to 15 s
Pika5 s and 10 s, 20-25 s on Pikaframes
PixVerse1 to 15 s on version 6
Runway5 to 15 s on Gen-4.5
SceneMixerThirty seconds on the Wan tiers, fifteen on H3
ViduNot stated
Wan 3.02 to 30 s

Inclusion rule. Every generator this register carries, on the one column this pair disagrees about. Rows that state nothing keep their place and are marked hollow. Order. Alphabetical by generator name.

Why these two. Twenty times apart in the same unit, and the comparison is meaningless until the mechanism is printed beside it. This pair is why these sheets refuse to sort the length column.

5Sources

Read from the Hedra Character-3 model page and the Wan3.0 API reference, both checked 2026-09-22. The column is described on duration, and each entry has its own cell page at Hedra and Wan 3.0, and the rule that keeps surfaces and generators apart is on method. Nearby pairs: CogVideoX and HeyGen, Wan 3.0 and Veo.