Fast, Lite and Mini: what you give up on a cheaper tier

A cheaper tier is only useful if you know what it costs you. On this roster the naming doesn't tell you, and in one case it points the wrong way.

What do the three words actually mean?

Nothing consistent, and that's the finding.

There's no roster-wide convention. Fast on Veo means the same capability rendered quicker. Fast on Seedance means a lower resolution ceiling. Lite on Veo means dropped features. Mini on Seedance means a lower resolution ceiling — the same thing Fast means there.

So the word is a hint rather than a specification, and you have to read the settings either way.

Fast on Veo keeps everything. Fast on Seedance drops resolution. The same word, two different trades.

The Veo tiers

Three models, and the naming is actively misleading about which one costs you more.

Read that column by column and the naming inverts.

Veo 3.1 Fast gives up nothing the picker exposes. Same eight-second lock, same three resolution tiers including 4K, same three references, same everything. Its description says "faster Veo, same sound and scene extension," which is accurate — it's a speed tier and nothing else.

Veo 3.1 Lite is the one that actually costs you. No references section at all, which means no character consistency across shots and no keyframe-plus-reference workflow. And no 4K.

The one thing Lite gains is flexibility: a 4–8 second slider rather than a lock. So it's the only Veo model where you choose the duration.

Veo 3.1 Fast costs you nothing. Veo 3.1 Lite costs you references and 4K. The names suggest the reverse.

Where this falls short. "Gives up nothing the picker exposes" is the honest limit of that claim. Render speed isn't in the picker, and neither is output quality. Both could differ and we haven't measured either.

The Seedance tiers

Four models, and two of them are hard to tell apart.

Three things fall out of that.

Seedance 2 Fast and Seedance 2 Mini have identical settings. Same duration, same two resolution tiers, same 15 references, same seven aspect ratios plus auto, same audio toggle, same frames. Every figure the picker exposes is the same.

Only the descriptions differ — "quicker Seedance passes for iteration" against "lightweight motion for rapid drafts." So the distinction is presumably speed or cost rather than capability, and the picker doesn't tell you which.

Seedance 2.0 is the only Seedance with 4K. Not 2.5, the flagship. That's worth knowing if a deliverable specifies it.

All four keep the same reference and aspect ratio behaviour except 2.5's fifty-reference ceiling. So dropping to a cheaper Seedance tier costs you resolution and duration, never framing or reference capacity.

What do the published costs suggest?

The render costs panel gives figures for three of the four Seedance tiers, and the spread is significant.

For a five-second clip at 720p: Seedance 2 Mini at 129 credits, Seedance 2 Fast at 202, Seedance 2.0 at 263. At 480p it's 60, 94 and 122.

So Mini is cheaper than Fast despite identical settings, which suggests the difference between them is priced rather than specified. And 2.0 costs roughly double Mini for the same reference ceiling and the same aspect ratios — you're paying for the 1080p and 4K tiers and nothing else the picker shows.

Where this falls short. Those figures come from a published pricing panel rather than the picker, and that panel is incomplete — Seedance 2.5 isn't on it at all. So treat the ratios as indicative and check the cost shown before you render, which is the authoritative number.

What about the other Fast models?

Two more on the roster, and neither has a full-price sibling.

LTX-2.3 Fast has no LTX-2.3 to compare it to. It's Fast as a product name rather than a tier, and its settings are odd for a draft model — four resolution tiers including 1440p and 4K, which is more range than most delivery models, against a two-reference ceiling.

Vidu Q3-Turbo is the same situation. Turbo with no base model, and its identity is keyframing rather than speed — "first and last frame, up to sixteen seconds."

Kling 3.0 Turbo does have a sibling in Kling 3.0 Omni, and the trade there is clear: Omni gives you seven references and an audio toggle, Turbo gives you one reference and audio always on. Same duration, same resolutions, same aspect ratios.

When should you actually use a cheaper tier?

Three cases, and one of them is the reason these tiers exist.

All settings read from Hexcoded's Creative Studio picker in Video → Create mode in September 2026. Credit figures are from Hexcoded's published render costs panel, which is incomplete. Settings and costs change with model versions. The cost shown before you render is the authoritative figure.

More on model capability, access and rights in Models.