AI Video Generation Pricing Compared (October 2026): What a Second Really Costs

There are two ways to buy AI video in 2026: per second from an API, or per credit from a consumer app. The second is convenient and almost always more expensive, because the credit-to-second rate is rarely published and changes without notice. Here are the numbers.

Per-second API prices, read 4 September 2026 (updated 18 October)

Every figure below was pulled from the live fal.ai endpoint on 4 September; the OmniHuman row and the October update at the end were re-read on 18 October. Prices move; the endpoint IDs are in the developer block at the bottom so you can re-check.

The August version of this table spanned 2.6×. The September one spans 15×, and both new ends are informative: Wan 3.0 Prime replaced Wan 2.7 at half the rate, while Seedance 2.5 bills in video tokens ($0.0214 per 1,000) that work out near half a dollar a second — and refuses real human faces, so the price buys stock footage. See the autumn model round-up for the full release list.

What that means for a real Instagram Reel

A typical Reel worth recreating is 8–15 seconds. At the prices above, the raw model cost of one vertical clip lands between $0.80 and $2.60. Add analysis, storage and retries and the true cost of a finished short video is roughly $1–4 in compute.

Consumer tools sell the same models at $9–49 per month with a credit pool, which is fine if you publish daily and terrible if you publish twice a month. The break-even is easy to compute once you know the per-second rate — which is the whole reason vendors avoid printing it.

Three pricing traps

Resolution and audio multipliers on credit platforms. Several per-second API rates above are flat, but not all: HappyHorse 1.1 bills about double at its 1080p default, and Wan 3.0 Prime's $0.05 is a base unit that rises with resolution. Credit-based consumer plans are different: on Kling's subscription, a Kling 3.0 clip runs about 6 credits per second at 720p without audio and about 12 with 1080p and native audio. Two toggles, double the burn, and the rate is not shown next to the toggles.

Endpoints with no published price. Some models have no per-output rate in fal's pricing API — it answers with the platform's GPU-fallback rate, which is what it returns for anything missing from its pricing table. You cannot quote such a job to a user before it runs. We exclude these from production routing for exactly that reason. That excluded most of a release cycle in late August 2026; by 18 October fal lists base rates for most of it (see the update below), though not the resolution multipliers.

Expiring top-ups. Credit packs that expire in 90 days are now an industry norm. If your usage is bursty, a meaningful share of what you buy is never spent.

Cheapest is not the answer, matched is

Wan 3.0 Prime at its $0.05 base rate looks like the cheapest full-quality option on the list (at 720p and above the reported rate is higher; see the update below) — the FLUX.3 draft row is cheaper, but a draft tier is for proving a composition, not for publishing — and it animates a still — it has nothing to say about an existing clip. Wan 2.7 edit-video at $0.10/s is twice the base price and the only one of the two that keeps your background (though we tested it and do not route to it). For a talking-head clip, both are the wrong tool: neither has anything to say about lip-sync from a script. HeyGen translate at $3.00/min looks expensive next to $0.14/s until you notice it does translation, timbre preservation and lip-sync in one call, which would otherwise be three billed steps.

The cost that matters is not $/second. It is dollars per usable clip, and that number is dominated by how many times you re-run a job because the model was a bad fit.

October 2026 update

Prices read on 18 October 2026 from fal's pricing data and, where fal shows only a base unit, from third-party trackers (marked "reported"):

Two caveats worth carrying. Base rates for the H3 Max and Wan lines understate the price at 1080P, and fal publishes no multiplier. And a per-second price tells you nothing about identity: none of these models has been certified by anyone for keeping a face stable, which is why we test before we route. For one-page context, see AI video models in October 2026.

How to do this without code: Per-second model prices are the wrong unit if you want one finished video, so the product quotes the whole thing in credits first. In AutoHustle this is the autopilot: paste the Reel link, add one photo of yourself and confirm the plan (about 4 credits per second of finished video, so 60 credits for a full 15-second call; analysing the Reel is free). See How to put your face in any Reel and Recreate an Instagram Reel.

Related reading: HappyHorse Video Edit on fal: What It Costs to Replace a Person in a Clip, and the Limits (October 2026) · Kling 3.0 vs HappyHorse 1.1 for Character Replacement (August 2026)

For developers: the fal endpoints behind this article

Prices in the article are per-second or per-unit rates from the fal.ai pricing API on the dates stated; re-check them before you build, because endpoints reprice without notice. The IDs:

Model	fal endpoint
FLUX.3 draft (t2v and i2v)	blackforestlabs/flux-3/*/draft
Wan 3.0 Prime	alibaba/wan-3.0-prime/*
One-to-All Animation 14B	fal-ai/one-to-all-animation/14b
FLUX.3 text-to-video	blackforestlabs/flux-3/text-to-video
Wan 2.7 edit-video	fal-ai/wan/v2.7/edit-video
Kling 3.0 Turbo Standard image-to-video	fal-ai/kling-video/v3/turbo/standard/image-to-video
Kling 3.0 Standard Motion Control	fal-ai/kling-video/v3/standard/motion-control
Kling 3.0 Pro image-to-video	fal-ai/kling-video/v3/pro/image-to-video
HappyHorse 1.1	alibaba/happy-horse/v1.1/*
OmniHuman 2.0	bytedance/omnihuman/v2
Kling 3.0 Pro Motion Control	fal-ai/kling-video/v3/pro/motion-control
MiniMax H3 reference-to-video	minimax/h3/reference-to-video
Kling Native 4K image-to-video	fal-ai/kling-video/v3/4k/image-to-video
Seedance 2.5	bytedance/seedance-2.5/*
HeyGen Translate v2 (precision)	fal-ai/heygen/v2/translate/precision