AI video generation API pricing
18 video models from 6 companies, priced per second of generated video. At 720p the cheapest is Grok Imagine Video 1.5 Lite at $0.03 a second; Google’s Veo 3.1 costs $0.20–$0.40 a second, depending on audio. Each price is checked on the company’s own pricing page, 6 Nov 2026.
Video generation API prices
Cheapest first
| Model | Resolution compared | Per second | Per minute | Other resolutions | Source |
|---|---|---|---|---|---|
 Grok Imagine Video 1.5 LitexAI · grok-imagine-video-1.5-lite | 720p | $0.03 | $1.80 | 480p $0.02 · 1080p $0.14 | Source for Grok Imagine Video 1.5 Lite: xAI API pricing (opens in a new tab) |
 Veo 3.1 LiteGoogle | 720p | $0.03–$0.05No audio to With audio | $1.80up to $3.00 | 1080p $0.05–$0.08 | Source for Veo 3.1 Lite: Google Cloud generative AI pricing (opens in a new tab) |
 MiniMax H3 LiteMiniMax · MiniMax-H3-Lite | 768p | $0.04 | $2.40 | 480p $0.025 | Source for MiniMax H3 Lite: MiniMax pricing (opens in a new tab) |
Gen-4 TurboRunway · gen4_turbo | One rate for all outputs | $0.05 | $3.00 | — | Source for Gen-4 Turbo: Runway API pricing (opens in a new tab) |
| FLUX 3 Video DraftBlack Forest Labs | HD | $0.06 | $3.60 | — | Source for FLUX 3 Video Draft: Black Forest Labs pricing (opens in a new tab) |
MiniMax H3 HighspeedMiniMax · MiniMax-H3-Highspeed | 768p | $0.06 | $3.60 | 480p $0.04 | Source for MiniMax H3 Highspeed: MiniMax pricing (opens in a new tab) |
Gemini Omni 1.2 FlashGoogle · gemini-omni-1.2-flash | 720p | $0.07maker’s estimate | $4.20 | 360p $0.023 · 1080p $0.104 · 4K $0.209 | Source for Gemini Omni 1.2 Flash: Gemini API pricing (opens in a new tab) |
Grok Imagine VideoxAI · grok-imagine-video | 720p | $0.07 | $4.20 | 480p $0.05 | Source for Grok Imagine Video: xAI API pricing (opens in a new tab) |
MiniMax H3MiniMax · MiniMax-H3 | 768p | $0.08 | $4.80 | 2K $0.13 | Source for MiniMax H3: MiniMax pricing (opens in a new tab) |
MiniMax H3 MaxMiniMax · MiniMax-H3-Max | 768p | $0.08 | $4.80 | 480p $0.05 | Source for MiniMax H3 Max: MiniMax pricing (opens in a new tab) |
| Veo 3.1 FastGoogle | 720p | $0.08–$0.10No audio to With audio | $4.80up to $6.00 | 1080p $0.10–$0.12 · 4K $0.25–$0.30 | Source for Veo 3.1 Fast: Google Cloud generative AI pricing (opens in a new tab) |
Gen-4.5Runway · gen4.5 | One rate for all outputs | $0.12 | $7.20 | — | Source for Gen-4.5: Runway API pricing (opens in a new tab) |
Grok Imagine Video 1.5xAI · grok-imagine-video-1.5 | 720p | $0.14 | $8.40 | 480p $0.08 · 1080p $0.25 | Source for Grok Imagine Video 1.5: xAI API pricing (opens in a new tab) |
 HappyHorse 1.0Alibaba · happyhorse-1.0-t2v | 720p | $0.14 | $8.40 | 1080p $0.24 | Source for HappyHorse 1.0: Alibaba Cloud Model Studio pricing (opens in a new tab) |
HappyHorse 1.1Alibaba · happyhorse-1.1-t2v | 720p | $0.14 | $8.40 | 480p $0.07 · 1080p $0.18 | Source for HappyHorse 1.1: Alibaba Cloud Model Studio pricing (opens in a new tab) |
| FLUX 3 VideoBlack Forest Labs | HD | $0.15 | $9.00 | Full HD $0.24 · 2K $0.28 · 4K $0.41 | Source for FLUX 3 Video: Black Forest Labs pricing (opens in a new tab) |
| Veo 3.1Google | 720p–1080p | $0.20–$0.40No audio to With audio | $12.00up to $24.00 | 4K $0.40–$0.60 | Source for Veo 3.1: Google Cloud generative AI pricing (opens in a new tab) |
Aleph 2.0Runway · aleph2 | One rate for all outputs | $0.28 | $16.80 | — | Source for Aleph 2.0: Runway API pricing (opens in a new tab) |
US dollar list prices before tax, checked 6 Nov 2026 on each company’s own pricing page. Prices are per second of output at the resolution shown (MiniMax’s standard is 768p; Black Forest Labs’ HD); a range shows the cheapest to the dearest setting, such as without and with audio. Per minute is 60 seconds at the cheapest setting.
Other video models through Runway’s API
| Model | Resolution compared | Per second | Per minute | Other resolutions | Source |
|---|---|---|---|---|---|
Kling 3.0Kuaishou · through Runway · kling3.0_standard, kling3.0_pro | One rate for all outputs | $0.09–$0.17Standard, no audio to Pro, audio | $5.40up to $10.20 | 4K $0.43 | Source for Kling 3.0: Runway API pricing (opens in a new tab) |
Wan 3Alibaba · through Runway · wan3 | 720p | $0.10 | $6.00 | 480p $0.05 · 1080p $0.20 | Source for Wan 3: Runway API pricing (opens in a new tab) |
Kling 4Kuaishou · through Runway · kling_4 | 720p | $0.157 | $9.42 | 1080p $0.235 | Source for Kling 4: Runway API pricing (opens in a new tab) |
Seedance 2.5ByteDance · through Runway · seedance2_5 | 720p | $0.30 | $18.00 | 480p $0.20 · 1080p $0.68 | Source for Seedance 2.5: Runway API pricing (opens in a new tab) |
Runway’s own rates for other companies’ models, at $0.01 a credit. Their makers may sell direct access at different prices; we list only prices we could check on a seller’s own page.
What each price includes
- Grok Imagine Video 1.5 Lite xAI
- Each input image costs $0.01.
- Veo 3.1 Lite Google
- Sold on Google Cloud (Vertex AI).
- MiniMax H3 Lite MiniMax
- Each input image costs $0.02.
- Gen-4 Turbo Runway
- 5 credits a second at $0.01 a credit.
- FLUX 3 Video Draft Black Forest Labs
- For quick drafts; HD only.
- MiniMax H3 Highspeed MiniMax
- Each input image costs $0.03.
- Gemini Omni 1.2 Flash Google
- Billed per token at $12 per 1M video output tokens; the per-second figures are Google's own (5,792 tokens a second at 720p; 1080p and 4K are upsampled). Up to 15 seconds with native audio. Gemini Omni 1.1 Flash costs $17.50 per 1M (about $0.10 a second) until 3 December 2026, when its requests move to 1.2 at 1.2's price. Gemini Developer API, paid tier.
- Grok Imagine Video xAI
- Also edits video: input video costs $0.01 a second and each input image $0.002.
- MiniMax H3 MiniMax
- Input video is billed at the same rate per second; each input image costs $0.04.
- MiniMax H3 Max MiniMax
- Takes image, video and audio references. Input video costs $0.143 a second at 768p; each input image $0.074.
- Veo 3.1 Fast Google
- Sold on Google Cloud (Vertex AI).
- Gen-4.5 Runway
- 12 credits a second at $0.01 a credit.
- Grok Imagine Video 1.5 xAI
- Text, image and audio to video. Each input image costs $0.01.
- HappyHorse 1.0 Alibaba
- Text to video, international deployment (Singapore). Only output is billed.
- HappyHorse 1.1 Alibaba
- Text to video, international deployment (Singapore). Only output is billed.
- FLUX 3 Video Black Forest Labs
- Text or image to video. Video to video costs $0.21 a second at HD, up to $0.45 at 4K.
- Veo 3.1 Google
- Sold on Google Cloud (Vertex AI).
- Aleph 2.0 Runway
- Video editing. 28 credits a second, with a 56-credit ($0.56) minimum per generation.
- Kling 4 Kuaishou
- List price without an input video; 20% off until 30 November 2026.
- Seedance 2.5 ByteDance
- Input and reference video cost half the output rate; 80-credit ($0.80) minimum per generation.
How video APIs charge
- Per second of output. Every model here bills by the length of the video it makes, so a 10-second clip costs twice a 5-second one, and drafts you throw away cost as much as the final cut.
- Resolution. Higher resolutions cost more. On the models here that offer both, 1080p costs 1.3 to 4.7 times the 720p rate, and 4K more again.
- Audio. Generated sound doubles Veo 3.1’s price, from $0.20 to $0.40 a second.
- Input video. Editing or extending a video can also bill the input footage: Grok Imagine Video, MiniMax H3 and MiniMax H3 Max do.
- Minimum charges. Aleph 2.0 and Seedance 2.5 have a minimum charge per generation.
- Tokens. Google bills Gemini Omni Flash per output token and publishes an approximate per-second price; the other models list a price per second directly.
Questions
What is the cheapest video generation API?
At 720p, Grok Imagine Video 1.5 Lite from xAI is the cheapest at $0.03 a second, or $1.80 for a minute of video. Quality and clip length vary a lot between models, so test on your own prompts.
How much does Veo 3.1 cost?
On Google Cloud (Vertex AI), Veo 3.1 costs $0.20 a second at 720p or 1080p without audio and $0.40 with audio, so an 8-second clip with sound costs $3.20. Veo 3.1 Lite costs $0.03–$0.05 a second at 720p and Veo 3.1 Fast costs $0.08–$0.10 a second at 720p.
What does a minute of AI video cost?
From $1.80 on Grok Imagine Video 1.5 Lite to $16.80 on Aleph 2.0, at the resolutions in the table. Most providers bill per second of output, so you pay for every draft as well as the final cut.
How is Gemini Omni Flash billed?
Billed per token at $12 per 1M video output tokens; the per-second figures are Google's own (5,792 tokens a second at 720p; 1080p and 4K are upsampled). Up to 15 seconds with native audio. Gemini Omni 1.1 Flash costs $17.50 per 1M (about $0.10 a second) until 3 December 2026, when its requests move to 1.2 at 1.2's price. Gemini Developer API, paid tier.
Can I get Kling or Seedance through an API?
Runway’s API sells both at its own rates: Kling 3.0 at $0.09–$0.17 a second, Kling 4 at from $0.157 a second at 720p and Seedance 2.5 at from $0.30 a second at 720p. Their makers may also sell direct access; we list only prices we could check on a seller’s own page.
When were these prices checked?
On 6 Nov 2026, on each company’s own pricing page, linked from every row. Prices are in US dollars before tax, per second of generated video.
More pricing
Sources: xAI API pricing, Google Cloud generative AI pricing, MiniMax pricing, Runway API pricing, Black Forest Labs pricing, Gemini API pricing, Alibaba Cloud Model Studio pricing. Checked 6 Nov 2026.