GPU.ai has a new look. The compute is the same.Read about the redesign  

model apis · video generation

H3 API

Run H3 by MiniMax through one OpenAI-compatible endpoint. Pay per use against your GPU.ai balance. No instance to rent, no deployment to manage.

Available now

H3 API pricing

Live catalog rates, billed per use against your GPU.ai balance. No subscriptions, no minimums.

TierRate
Serverless
Per second of video: $0.139Per clip (from): $0.696

rates are read hourly from the live model catalog and metered per request

Call H3 in one request

Any OpenAI SDK works: set the base URL to https://api.gpu.ai/v1 and use your GPU.ai API key.

curl
# Start a generation, then poll it until it completes.
curl https://api.gpu.ai/v1/videos \
  -H "Authorization: Bearer $GPUAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "gpuai/h3",
    "prompt": "A drone shot over a mountain lake at dawn"
  }'

curl https://api.gpu.ai/v1/videos/{video_id} \
  -H "Authorization: Bearer $GPUAI_API_KEY"

H3 on GPU.ai

Model idgpuai/h3
ModalityVideo generation
AuthorMiniMax
Context window
Streaming
Aliases

faq

H3 API questions

How much does the H3 API cost?

H3 on GPU.ai currently costs $0.139 (per second of video), $0.696 (per clip (from)). There are no subscriptions or minimums. Usage is billed against your GPU.ai balance as you go.

Is the H3 API OpenAI-compatible?

Yes. Point any OpenAI SDK at https://api.gpu.ai/v1 with a GPU.ai API key and request model "gpuai/h3". No proprietary client needed.

How does billing work?

Generation models bill per output (per image, or per second of video) at the listed rate, metered per request against your GPU.ai balance.