How much does AI video generation cost per second via API?

Through an API, AI video generation is priced per second of generated video on most models: the cost of a clip is the per-second rate times its length, and the model, resolution and sound set the rate. On the vendors' own pricing pages read on 2026-09-29, listed rates run from $0.042 a second (Kling 2.5 Turbo at 720p on Kling's API) to $1.50 a second (150 credits on Runway's API for seedance2 at 4K). Each completed take is billed, including the ones you throw away.

On Sume the billed rate (the provider's list price × 1.25) runs from $0.0125 a second on grok-imagine-video-1.5 to about $1.43 a second on seedance-2.5 at 1080p, plus a 5.5% agent fee by default; a 10-second 720p clip from a prompt or a photo costs $0.13 to $5.82.

The Sume rates below are computed from the provider list prices in Sume's pricing code and the × 1.25 billing rule in the Video generation docs, read on 2026-09-29. For live rates, GET /v1/videos/models lists each model's pricing_skus, rounded to four decimals.

How much does AI video cost per second on Sume?

Sume bills every video model at the provider's list price × 1.25 and reserves the amount from your workspace balance when you submit. Most models charge per second of output, and most of those charge more per second at a higher resolution. The Seedance models charge per 1,000 video tokens instead.

What do Runway, Kling, fal and Replicate charge per second?

Each vendor's own pricing page, read on 2026-09-29. The USD column is only filled where the page states the dollar amount or the credit price: Runway sells credits at $0.01 each, and Kling prints the dollar price beside each unit count. The same model can carry a different price on different APIs, so compare the model rows here with Sume's table above.

• Runway adds charges on some models: seedance2_5 bills input and reference video per second and has an 80 credit minimum per generation, seedance2_mini has a 64 credit minimum, and grok_imagine_1_5 adds 1 credit per image or audio reference, including an image-to-video start frame.
• Runway's ProRes and PNG-sequence outputs add 5 credits per second, and its 10-bit and HDR profiles add 20 (40 when the output is larger than 4 megapixels), on Gen-4.5 and Aleph 2.0. A Model Router generation is billed at the standard rate of the model the router selects.
• Sound: Runway lists veo3.1 at 20 credits per second without audio and 40 with audio, and Kling 3.0 at 1080p rises from 0.8 to 1.2 Units a second with native audio and no voice control. Veo is not in Sume's catalog.
• fal bills video models per second or per video, depending on the model, and lists Veo 3 at $0.4 a second.

Which video API fits which job?

Sume fits when you want several model families on one key and one USD balance, with each clip's price reserved on submit and refunded if the job fails. It does not fit when you need a model outside its catalog, such as Runway's Gen models or Veo, and every Sume rate is its provider's list price × 1.25 plus the agent fee. One API for Kling, Seedance and other video models covers what one bill looks like, and fal.ai and Replicate alternatives for video compares the platforms.

• Pick Runway's API if you need gen4.5, gen4_turbo or veo3.1, or ProRes and HDR outputs. See Does Runway have an API?
• Pick Kling's API if you only call Kling models, including Kling 3.0 Omni, O1 or 2.6, or want Kling 3.0 at 4K. See Kling 3.0 API.
• Pick fal if you want output-priced model APIs and your own deployments on GPUs billed by the hour.
• Pick Replicate if you run community or fine-tuned models billed by the time they run.
• Pick Sume if you call several of the model families in its table above from one backend and want the per-second rates readable from GET /v1/videos/models.

What does a usable clip cost after retries?

The listed rate prices one take. If you generate three takes for each clip you keep, a kept clip costs three times the take. On Sume a job that completes is charged whether or not you use the clip; only a failed job, or one canceled before capture, is refunded. The example below is 100 kept 8-second clips a month at 3 takes each, 2,400 generated seconds in all.

How are Seedance clips priced?

By video tokens, billed per 1,000, so the per-second price follows frame size: a 720p 16:9 second is 21,600 tokens, about $0.58 a second on seedance-2.5, and a 10-second 1080p 16:9 clip reserves $14.22 because 1080p has more pixels and its own per-token rate. Seedance tokens: how video tokens turn into dollars walks through the formula.

How much does a minute of AI video cost?

No single clip runs a minute. seedance-2.5 and wan-3.0 stop at 30 seconds, and every other model at 15 seconds or less, so a minute of footage is several clips; see AI video length limits by model. Sixty seconds of output costs $0.76 on grok-imagine-video-1.5 (four 15-second clips), $7.50 on wan-3.0 at 720p (two 30-second clips), and $22.50 on gemini-omni-flash-1.1 at 4K (six 10-second clips), before the fee.

Joining the clips into one MP4 with Timeline 1.0 is listed at $0.10 per output minute on API pricing; How to assemble a long-form video shows the render.

Does image-to-video cost the same as text-to-video?

Yes, for the same model, length, resolution, and aspect ratio. A model's rates in pricing_skus vary by resolution, or by sound on kling-3, and none of them depends on whether the clip starts from a prompt or from your photo as the first frame.

References are where extra cost can appear. For a reference video on the Seedance models, Sume's current estimate adds assumed input seconds to the token count; Why are AI videos so expensive? covers that and the other cost drivers.

What else changes what I pay?

• The agent fee: 5.5% of the model amount by default, explained in What is the Sume agent fee?
• Failed jobs: the reserved amount is refunded on failure, or on cancellation before capture. See Do failed AI video jobs cost money?
• sume/auto: Sume picks the model and does not disclose which. Read the job's usage.cost, the Sume billable amount, when it completes.
• Steps after generation: API pricing lists Video Upscale at $0.009 per video second, text to speech at $0.0475 per 1,000 characters, and music generation at $0.125 per audio.
• Checking first: Estimate AI video cost before you run lists the ways to see a price before paying, and AI video cost per run reads what a finished job cost.