Pricing

Pay as you go, per token, image or second of video. No subscriptions, no minimums, and no charge for failed generations.

Video

ModelPriceExample
Veo 3.1 Fast
ewest/veo-3.1-fast
720p: $0.10 per second
1080p: $0.12 per second
A 4 s 720p clip costs $0.40.
Seedance 2.5
ewest/seedance-2.5
480p: $0.2056 per second
720p: $0.4622 per second
1080p: $1.04 per second
A 5 s 480p clip costs $1.03.
MiniMax H3 Max
ewest/minimax-h3-max
480p: $0.05 per second
768p: $0.08 per second
A 5 s 480p clip costs $0.25.

Image

ModelPriceExample
FLUX.1 [schnell]
ewest/flux-schnell
Image: $0.003 per image
1,000 images cost $3.
Gemini 2.5 Flash Image
ewest/gemini-flash-image
Image: $0.039 per image
Prompt (input): $0.30 per million tokens
100 images cost about $3.90.

Language

ModelPriceExample
Gemini 2.5 Flash
ewest/gemini-flash
Input: $0.30 per million tokens
Output: $2.50 per million tokens
1,000 requests of 1,000 tokens in and 300 out cost about $1.05.
Claude Haiku 4.5
ewest/claude-haiku
Input: $1.00 per million tokens
Output: $5.00 per million tokens
1,000 requests of 1,000 tokens in and 300 out cost about $2.50.
Claude Sonnet 5.5
ewest/claude-sonnet
Input: $2.00 per million tokens
Output: $10.00 per million tokens
1,000 requests of 1,000 tokens in and 300 out cost about $5.

Questions

Are failed requests charged?
No. You pay only for successful generations; a failed video job is recorded at $0.
How is video billed?
Per generated second, at the price for the resolution you choose. The cost is shown on the job when it finishes.
How do I control spend?
Set a budget on each API key, and limit which models a key may use. Usage and cost are broken down by key and model in the console.
Are there minimums or subscriptions?
No. Prices are per unit with no monthly fee.