Original data

AI video API benchmark (November 2026): render time, success rate and price per second for every model

How long does an AI video API take, how often does it deliver, and what does a second of video cost? We measured 68,309 real video jobs that customers sent through the Revana API across 24 models, plus 148,607 image jobs. Here are the numbers, with charts and a CSV you can download and cite.

  • 34 smedian render, MiniMax H3 Max (768p), the fastest video model
  • 99.1%of 48,947 Seedance 2.5 jobs delivered, the most reliable model at volume
  • $0.03a second at 720p for Grok Imagine Video 1.5 Preview, the lowest 720p list price
  • 72%of customer video jobs went to Seedance 2.5

Коротко

  • Fastest: MiniMax H3 Max (34 s), MiniMax H3 (64 s), Grok Imagine Video 1.5 Preview (67 s), Gemini Omni 1.2 Flash (67 s), Gemini Omni (75 s). Median time from request to finished file, at each model's most used resolution.
  • Slowest per second of video: the Seedance models. Seedance 2.5 took a median 358 s at 720p for a median 9 second clip, about 41 seconds of waiting for each second of video. MiniMax H3 Max needed 5.
  • Most reliable: Seedance 2.5 delivered 99.1% of 48,947 jobs. Kling 2.5 Turbo Pro, Kling 4.0 Flash, Hailuo 02 Standard, Wan 2.7 delivered every job, on smaller volumes.
  • Least reliable: Gemini Omni 1.1 Flash (83.4% over the period, 92.1% in the last 7 days) and MiniMax H3 (84.8%, but 124 of 124 in the last 7 days).
  • Cheapest per second: Hailuo 02 at 512p ($0.02), then Grok Imagine Video 1.5 Preview at 720p ($0.03). Prices are Revana's list prices; see the pricing page for every rate.

Render time per model

Render time here is the wall-clock time from the moment the API accepted a request to the moment the finished file was stored and ready to download. It includes queueing and, when a route failed, the retry on the next route. So it is what a developer waiting on the API actually experiences, not a lab number.

0 s 225 s 450 s 675 s 900 s MiniMax H3 Max · 768p 34 s MiniMax H3 · 768p p90 1285 s → 64 s Grok Imagine 1.5 Prev. · 480/720p 67 s Gemini Omni 1.2 Flash · 720p 67 s Gemini Omni · 720p 75 s Veo 3.1 Lite · 720p 78 s Wan 2.7 · 720p 85 s Gemini Omni 1.1 Flash · 720p 89 s Veo 3.1 Fast · 720p 95 s Kling 4.0 Flash · 720p 100 s Hailuo 02 · 512p 108 s Kling 3.0 · 720p 117 s Kling 3.0 Omni · 720p 136 s Veo 3.1 · 1080p 155 s Seedance 2.0 Fast · 720p 160 s Wan 2.6 I2V Flash · 720p 161 s Seedance 2.0 Mini · 720p 163 s Wan 3.0 Prime · 720p 166 s Wan 3.0 · 720p 195 s Kling 2.5 Turbo Pro · std 260 s Hailuo 02 Standard · 768p 262 s Kling 4.0 · 720p 264 s Seedance 2.0 · 720p 332 s Seedance 2.5 · 720p 358 s
Bars: median (p50) render time. Lines: 90th percentile (p90), the time 9 in 10 jobs finished within (an arrow when it runs past the chart). Each model at its most used resolution; clip lengths are whatever customers asked for (median in the table below).

Clip length matters: a 15 second clip takes longer than a 5 second one on the same model. To compare fairly, we also divided each job's render time by the length of the video it produced:

0 s 15 s 30 s 45 s 60 s MiniMax H3 Max · 768p 5 s Gemini Omni 1.2 Flash · 720p 10 s Wan 2.7 · 720p 11 s MiniMax H3 · 768p 11 s Veo 3.1 Lite · 720p 12 s Veo 3.1 Fast · 720p 14 s Grok Imagine 1.5 Prev. · 480/720p 17 s Hailuo 02 · 512p 18 s Gemini Omni 1.1 Flash · 720p 18 s Wan 3.0 Prime · 720p 19 s Gemini Omni · 720p 19 s Kling 3.0 · 720p 24 s Kling 4.0 Flash · 720p 24 s Veo 3.1 · 1080p 28 s Seedance 2.0 Mini · 720p 31 s Seedance 2.0 Fast · 720p 32 s Wan 3.0 · 720p 33 s Kling 3.0 Omni · 720p 34 s Wan 2.6 I2V Flash · 720p 40 s Seedance 2.5 · 720p 41 s Hailuo 02 Standard · 768p 45 s Kling 4.0 · 720p 47 s Kling 2.5 Turbo Pro · std 52 s Seedance 2.0 · 720p 59 s
Median seconds of waiting per second of finished video. Lower is faster.

Two things stand out. First, the gap between the median and p90 is large on some models: MiniMax H3 had a median of 64 s but a p90 of 1285 s, because a few jobs waited a long time for capacity. If your product has a timeout, plan for the p90, not the median. Second, higher resolution is not always slower: for Seedance 2.5, 1080p jobs had a shorter median (237 s) than 720p (358 s) because people asked for shorter clips at 1080p (median 6 s against 9 s).

Render time by resolution

Render time by resolution
МодельРазрешениеDelivered jobsp50p90Median clip
Seedance 2.5480p19,455324 s556 s11 s
Seedance 2.5720p28,322358 s635 s9 s
Seedance 2.51080p723237 s426 s6 s
Gemini Omni 1.1 Flash720p3,35789 s160 s6 s
Gemini Omni 1.1 Flash1080p1,284148 s224 s6 s
Gemini Omni 1.1 Flash4K90263 s293 s8 s
Wan 3.0480p393158 s285 s5 s
Wan 3.0720p2,348195 s346 s5 s
Wan 3.01080p136241 s862 s6 s
Seedance 2.0480p140299 s446 s9 s
Seedance 2.0720p2,444332 s501 s5 s
Kling 3.0720p1,699117 s201 s5 s
Kling 3.01080p516150 s294 s4 s
Gemini Omni 1.2 Flash720p98767 s116 s8 s
Gemini Omni 1.2 Flash1080p126100 s160 s6 s
Gemini Omni 1.2 Flash4K32207 s242 s8 s
Seedance 2.0 Mini480p122133 s173 s5 s
Seedance 2.0 Mini720p921163 s226 s4 s
Kling 3.0 Omni720p403136 s176 s3 s
Kling 3.0 Omni1080p121141 s441 s4 s
Kling 4.0720p376264 s514 s5 s
Kling 4.01080p161293 s430 s5 s
Seedance 2.0 Fast480p235140 s186 s5 s
Seedance 2.0 Fast720p223160 s221 s5 s
Wan 3.0 Prime720p138166 s273 s8 s
Wan 3.0 Prime1080p119140 s265 s4 s
Veo 3.1 Fast720p8195 s101 s6 s
Veo 3.1 Fast1080p118136 s157 s4 s
MiniMax H3 Max480p4223 s26 s7 s
MiniMax H3 Max768p12634 s62 s5 s
MiniMax H3 Max1080p3029 s44 s15 s
Veo 3.1 Lite720p8778 s102 s8 s
Veo 3.1 Lite1080p40136 s175 s6 s

Success rate per model

A job counts as delivered when the API returned a finished video. Everything else counts as a failure, including content refusals (the model declined the prompt or the picture) and temporary outages. On Revana a failed job costs nothing, but it still costs you time.

0% 25% 50% 75% 100% Kling 2.5 Turbo Pro 100% Kling 4.0 Flash 100% Hailuo 02 Standard 100% Wan 2.7 100% Seedance 2.5 99.1% MiniMax H3 Max 99.0% Seedance 2.0 Mini 99.0% Wan 3.0 Prime 98.9% Hailuo 02 98.7% Kling 3.0 98.4% Veo 3.1 Fast 98.2% Wan 3.0 98.1% Kling 4.0 97.5% Seedance 2.0 97.2% Wan 2.6 I2V Flash 97.0% Veo 3.1 96.6% Seedance 2.0 Fast 96.0% Kling 3.0 Omni 94.4% Grok Imagine 1.5 Prev. 94.2% Gemini Omni 1.2 Flash 93.0% Veo 3.1 Lite 92.0% Gemini Omni 85.7% MiniMax H3 84.8% Gemini Omni 1.1 Flash 83.4%
Share of jobs delivered, 20 October to 15 November 2026. The table below adds the last 7 days and the rate without content refusals is in the CSV.

Most failures came in bursts, not evenly: Gemini Omni 1.1 Flash delivered 83.4% over the whole period but 92.1% in the last 7 days, and most of its failures were temporary capacity errors from the model's servers. Most of MiniMax H3's failures were jobs that waited too long for capacity in its first weeks; in the last 7 days it delivered all 124 jobs. Content refusals were a small share almost everywhere: for Seedance 2.5 they explain about half of the failures.

When a model fails, the Revana router retries the job on another route that serves the same model, so the numbers above already include that fallback. Read more in AI video API error codes.

Цена за секунду

Every video model on Revana is billed per second of output. To compare models with a per-job fee fairly, the chart spreads the fee over a 5 second clip. Prices are list prices for a 5 second clip with sound where the model charges extra for sound.

0¢ 10¢ 20¢ 30¢ 40¢ Hailuo 02 · 512p 2.0¢ Grok Imagine 1.5 Prev. · 480/720p 3.0¢ Hailuo 02 Standard · 768p 4.0¢ Seedance 2.0 Mini · 720p 5.0¢ MiniMax H3 · 768p 5.0¢ Wan 2.6 I2V Flash · 720p 5.0¢ Veo 3.1 Lite · 720p 5.0¢ Wan 3.0 · 720p 7.0¢ Gemini Omni 1.1 Flash · 720p 7.1¢ Gemini Omni 1.2 Flash · 720p 7.2¢ MiniMax H3 Max · 768p 8.0¢ Kling 2.5 Turbo Pro · std 8.0¢ Wan 2.7 · 720p 9.0¢ Veo 3.1 Fast · 720p 10¢ Kling 4.0 Flash · 720p 10¢ Gemini Omni · 720p 10¢ Seedance 2.0 Fast · 720p 11¢ Kling 3.0 · 720p 13¢ Kling 3.0 Omni · 720p 13¢ Wan 3.0 Prime · 720p 13¢ Kling 4.0 · 720p 14¢ Seedance 2.0 · 720p 15¢ Seedance 2.5 · 720p 21¢ Veo 3.1 · 1080p 40¢
List price per second of a 5-second clip at each model's most used resolution, in US cents. Prices change; the table below always shows the current catalogue price.

The cheapest model is not always the best value. A model that is twice the price but delivers on the first try and finishes in half the time can cost less in your product. Compare the price per second with the success rate and the p90 before you choose. For more ways to spend less, see how to cut AI video API costs and the cheapest AI video generation API.

Every video model in one table

Sorted by number of jobs. Each model's page has its full price list, inputs and code; the try link runs it in your browser.

Render time, success rate and list price per AI video model
МодельRes.ЗадачDeliveredПоследние 7 днейp50p90Median clipWait per video secondList price / sКлип 5 сTry
Seedance 2.5720p48,94799.1%99.4% (30,134)358 s635 s9 s41.4 s$0.21$1.05Try free
Gemini Omni 1.1 Flash720p5,67983.4%92.1% (4,392)89 s160 s6 s18.2 s$0.035 + $0.18 a job$0.355Try free
Wan 3.0720p2,93298.1%97.9% (1,887)195 s346 s5 s32.5 s$0.07$0.35Try free
Seedance 2.0720p2,67397.2%97.4% (2,584)332 s501 s5 s58.9 s$0.15$0.75Try free
Kling 3.0720p + sound2,25298.4%99.1% (1,488)117 s201 s5 s23.7 s$0.13$0.65Try free
Gemini Omni 1.2 Flash720p1,25093.0%93.2% (1,203)67 s116 s8 s9.8 s$0.072$0.36Try free
Seedance 2.0 Mini720p1,05499.0%99.8% (814)163 s226 s4 s31.1 s$0.05$0.25Try free
Kling 3.0 Omni720p + sound55594.4%94.5% (543)136 s176 s3 s34.3 s$0.13$0.65Try free
Kling 4.0720p55197.5%97.7% (352)264 s514 s5 s46.9 s$0.135$0.675Try free
Seedance 2.0 Fast720p47796.0%94.0% (316)160 s221 s5 s32.2 s$0.11$0.55Try free
MiniMax H3768p35684.8%100.0% (124)64 s1285 s5 s11.4 s$0.05$0.25Try free
Wan 3.0 Prime720p26498.9%98.7% (229)166 s273 s8 s18.5 s$0.13$0.65Try free
Veo 3.1 Fast720p + sound22498.2%97.8% (178)95 s101 s6 s13.7 s$0.10$0.50Try free
MiniMax H3 Max768p20099.0%98.6% (144)34 s62 s5 s4.8 s$0.08$0.40Try free
Wan 2.6 I2V Flash720p + sound16697.0%97.0% (165)161 s447 s4 s39.8 s$0.05$0.25Try free
Veo 3.11080p + sound14896.6%100.0% (73)155 s236 s6 s27.6 s$0.40$2.00Try free
Veo 3.1 Lite720p13892.0%90.7% (118)78 s102 s8 s12.1 s$0.05$0.25Try free
Kling 2.5 Turbo Prostd135100.0%100.0% (135)260 s767 s5 s51.7 s$0.08$0.40–
Hailuo 02512p7998.7%–108 s226 s6 s17.8 s$0.02$0.10Try free
Kling 4.0 Flash720p55100.0%100.0% (55)100 s181 s5 s24.2 s$0.10$0.50–
Grok Imagine Video 1.5 Preview480p/720p5294.2%94.2% (52)67 s87 s4 s16.8 s$0.03$0.15–
Hailuo 02 Standard768p46100.0%100.0% (46)262 s321 s6 s44.5 s$0.04$0.20–
Wan 2.7720p41100.0%100.0% (39)85 s150 s8 s10.8 s$0.09$0.45Try free
Gemini Omni720p3585.7%–75 s86 s4 s18.7 s$0.10$0.50Try free

What developers actually generate

0% 18% 35% 52% 70% Seedance 2.5 72% Gemini Omni 1.1 Flash 8.3% Wan 3.0 4.3% Seedance 2.0 3.9% Kling 3.0 3.3% Gemini Omni 1.2 Flash 1.8% Seedance 2.0 Mini 1.5% Kling 3.0 Omni 0.8% Everything else 4.3%
Share of 68,309 customer video jobs on the 24 models in this benchmark.

Seedance 2.5 carried 72% of all video jobs, even though it is one of the slower and pricier models per second. Developers pick it for quality, long clips (up to 30 seconds), sound, and reference images and video. The second most used model, Gemini Omni 1.1 Flash, is the opposite trade: fast and cheap per second, with a per-job fee.

Bonus: image models

The same measurement for the 14 image models with 30 or more jobs. Nano Banana Pro alone had 74,778 jobs; the fastest was Seedream 5.0 Flash at a median 22 s.

Render time, success rate and price per AI image model
МодельРазмерЗадачDeliveredp50p90List price / imageTry
Nano Banana Pro2K74,77898.9%51 s135 s$0.04Try free
GPT Image 22K36,96899.5%48 s152 s$0.03Try free
GPT Image 2.5 Sunburst2K18,19699.6%35 s66 s$0.02Try free
Seedream 5.0 Pro2K4,20699.6%108 s157 s$0.04Try free
Nano Banana 2.12K3,70998.7%43 s81 s$0.03Try free
Nano Banana 22K3,44599.0%31 s56 s$0.03Try free
GPT Image 2.5 Flare2K2,71198.6%46 s101 s$0.02Try free
GPT Image 2.6 Sunburst2K1,77999.1%50 s220 s$0.02Try free
GPT Image 2.6 Flare2K95298.2%36 s85 s$0.02Try free
Nano Banana 2 Lite1K88397.6%37 s79 s$0.02Try free
Seedream 5.52K52999.8%35 s61 s$0.055Try free
Seedream 5.0 Flash2K214100.0%22 s44 s$0.02Try free
Seedream 4.52K12794.5%35 s76 s$0.04Try free
Flux 3 Klein1K110100.0%26 s45 s$0.01Try free

How we measured

  • Source: every API and dashboard job customers sent to Revana between 20 October and 15 November 2026. Our own team and test accounts are left out.
  • Models: every video and image generation model with at least 30 jobs. Talking-avatar, lip-sync, restyle and upscaling tools are left out because they take audio or a source clip and are not comparable. Render times are shown per resolution only where at least 30 jobs finished at that resolution.
  • Render time: request accepted to file ready, for delivered jobs. p50 is the median; p90 is the time 9 in 10 jobs finished within.
  • Success rate: delivered jobs divided by all finished jobs (delivered plus failed). Cancelled and still-running jobs are not counted.
  • Prices: Revana list prices from our model catalogue as of November 2026. Every model has one price whichever route serves it.
  • Inputs vary: jobs mix text to video, image to video and reference jobs, and every clip length customers chose. That is the point: this is what real traffic looks like, not one test prompt.

You are free to use these numbers. Please link to this page as the source. We update it every month with the latest 30 days.

To run any of these models yourself, read the API docs or get an API key. One key and one request format work for every model in the table.

Частые вопросы

Which AI video API is the fastest?

In our data from 20 October to 15 November 2026, MiniMax H3 Max had the shortest median render time: 34 s at 768p for a median 5 second clip, from the moment the API accepted the request to the moment the file was ready. MiniMax H3 (64 s) and Grok Imagine Video 1.5 Preview (67 s) were next.

Which AI video model is the most reliable?

At volume, Seedance 2.5: 99.1% of 48,947 jobs were delivered (99.5% not counting content refusals). 4 smaller models delivered every job, the largest being Kling 2.5 Turbo Pro (135 of 135).

What is the cheapest AI video API per second?

On Revana in November 2026, Hailuo 02 at 512p costs $0.02 a second, and Grok Imagine Video 1.5 Preview costs $0.03 a second at 720p. Seedance 2.5 is $0.09 at 480p and $0.21 at 720p, with sound.

How long does Seedance 2.5 take to generate a video?

The median Seedance 2.5 job at 720p took 358 s and 9 in 10 finished within 635 s, for a median clip of 9 seconds. At 480p the median was 324 s.

How was render time measured?

From the time the API accepted the request to the time the finished file was stored and ready to download. It includes any queueing and any retry on a second route, so it is what a customer actually waited.

Can I use this data?

Yes. Download the CSV and cite this page as the source. We refresh the same page every month.

Try it first: run Seedance 2.5 free, MiniMax H3 Max free or Kling 3.0 free in your browser. Your first request is free once you save a card at sign-up (saving it charges nothing).

Every model in this benchmark, one API key

Call all 24 video models with the same key and request format, with automatic fallback to another route if one fails. Prepaid credit from $10.