Phosor API Quick Start
Choose your model guide first: MiniMax H3 Turbo or MiniMax H3. Each guide contains its own three model IDs and complete text, image and reference examples. This quick start explains the shared submit → status → result workflow.
API version: v1.2.4
Generate a video with MiniMax H3 using the Phosor API. The same asynchronous submit, status and result endpoints also support image generation.
API base URL: https://phosor.ai/api/v1
Authentication: send your API key in the X-API-Key header.
1. Get an API key
Sign in to Phosor, open API Keys, and create a key. Keep it on your server or in an environment variable; do not expose it in browser code. Add credits before submitting generation requests. Review current pricing before generating.
2. Submit a MiniMax H3 request
This example requests a five-second text-to-video generation at 480p. Set
PHOSOR_API_KEY to your API key before running the command.
curl --fail-with-body https://phosor.ai/api/v1/inference/submit \
-H "X-API-Key: $PHOSOR_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "minimax/h3/text-to-video",
"prompt": "A slow camera move across a sunlit coastal landscape, gentle waves and natural light",
"resolution_tier": "480p",
"aspect_ratio": "16:9",
"duration": 5
}'An accepted response includes a request ID:
{
"request_id": "your-request-id",
"status": "queued"
}Save request_id for the next steps. A browser or network timeout does not
prove submission failed. Check request history
before submitting again, to avoid creating duplicate paid requests.
3. Check the request status
Replace YOUR_REQUEST_ID with the ID returned by the submit endpoint.
curl --fail-with-body \
"https://phosor.ai/api/v1/inference/status/YOUR_REQUEST_ID?logs=false" \
-H "X-API-Key: $PHOSOR_API_KEY"Check periodically until status is completed or failed. Do not submit a
new generation request while waiting. See the status reference
for response fields and error handling.
4. Retrieve the video
Once the request is completed:
curl --fail-with-body \
https://phosor.ai/api/v1/inference/result/YOUR_REQUEST_ID \
-H "X-API-Key: $PHOSOR_API_KEY"The generated video URL is returned in data.video.url:
{
"request_id": "your-request-id",
"data": {
"video": {
"url": "https://example.com/generated-video.mp4"
}
}
}For an image model, read data.image.url instead. See the
result reference for the complete contract.
MiniMax H3 modes
| Mode | Standard model ID | Inputs |
|---|---|---|
| Text to video | minimax/h3/text-to-video | prompt |
| Image to video | minimax/h3/image-to-video | prompt, image_url; optional end_image_url |
| Reference to video | minimax/h3/reference-to-video | prompt, and reference images or reference videos |
Turbo variants use the same model ID with /turbo appended. H3 supports
480p and 768p, 4–15 second output durations, and fixed 24fps output. It does
not use Wan frame-count, Steps or Guidance controls. Reference input limits and
charges depend on the selected mode and resolution.
For image-to-video, upload an image or
import an image URL first. For multi-reference
generation, see the supported reference_image_urls, reference_video_urls
and reference_audio_urls parameters in the
submit reference.
Continue building
- Model API overview: endpoint workflow and public model IDs.
- Submit reference: generation parameters, supported models and examples.
- Common API contract: authentication, errors and asynchronous requests.
- LoRA integration: use custom models with supported workflows.
- API index: all available endpoint references.
- Pricing: output billing units and reference input charges.