Text to Video
kimi-h3
Kimi H3 text-to-video and image-to-video generation via the configured gputw provider
Commercial useAsync APIfrom 200 Credits
API reference
Submit this model to the capability endpoint, then poll the returned job until it is terminal. Only the parameters below are accepted.
Required headers
Authorization: Bearer $ARTANY_API_KEY Content-Type: application/json Idempotency-Key: unique-request-id
Use a new 8–128 character Idempotency-Key for each logical generation. Retrying the same request with the same key will not double-charge.
Request body
{
"model": "kimi-h3",
"prompt": "A cinematic slow orbit around a mountain lake at sunrise",
"duration": 5,
"resolution": "768p",
"aspect_ratio": "16:9"
}Parameters
| Field | Type | Required | Description | Default | Constraints |
|---|---|---|---|---|---|
| model | string | Yes | Stable Artany public model ID. | kimi-h3 | This exact public model ID. |
| seed | integer | No | Optional generation seed. The provider default is 42 when omitted. | — | Range: 0–2147483647 |
| prompt | string | Yes | Text instruction describing the scene, motion, and camera behavior. | — | Length: 1–3500 |
| duration | integer | No | Generated video duration in seconds. | 5 | Allowed: 5, 6, 7, 8, 9, 10 |
| image_urls | array (string items) | No | Optional first-frame image. Providing one image selects image-to-video generation. | — | Items: 0–1 |
| resolution | string | No | Text-to-video output canvas resolution tier. The value controls the canvas short edge before 32-pixel alignment. Do not send this field with image_urls. | — | Allowed: 480p, 768p |
| aspect_ratio | string | No | Text-to-video output aspect ratio. It is combined with resolution to derive the private provider width and height. Do not send this field with image_urls. | — | Allowed: 1:1, 2:3, 3:2, 4:3, 3:4, 4:5, 5:4, 16:9, 9:16 |
Accepted response
The submit endpoint returns HTTP 202 with a normalized job. Poll GET /v1/jobs/{jobId} until the status is terminal.
{
"id": "job_…",
"object": "generation.job",
"model": "kimi-h3",
"status": "queued",
"progress": 0,
"output": null,
"error": null,
"billing": {
"estimated_credits": "200",
"frozen_credits": "…",
"charged_credits": "0",
"state": "frozen",
"price_version": "2026-07-18.picverse-parity"
},
"created_at": "2026-07-21T00:00:00.000Z",
"completed_at": null
}Lifecycle and failures
- queued / running — wait and poll again.
- succeeded — read output URLs from
output; the actual charge is settled. - failed / canceled — inspect
error; frozen Credits are released. - For a schema error, the API returns an error object and identifies the invalid field in
paramwhen applicable.
Input
Runs are billed to your Credits at the listed price — same engine as the public API.
Output
Run the model to see results here
Pricing details
Fixed catalog pricing — the listed price is exactly what a run settles for. Version 2026-07-18.picverse-parity.
| Model | duration | Price |
|---|---|---|
| kimi-h3 | 5 | 200Credits · ~$2.0000 |
| kimi-h3 | 6 | 240Credits · ~$2.4000 |
| kimi-h3 | 7 | 280Credits · ~$2.8000 |
| kimi-h3 | 8 | 320Credits · ~$3.2000 |
| kimi-h3 | 9 | 360Credits · ~$3.6000 |
| kimi-h3 | 10 | 400Credits · ~$4.0000 |
Integrate in minutes
curl -X POST "https://api.artany.ai/v1/videos/generations" \
-H "Authorization: Bearer $ARTANY_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: unique-request-id" \
-d '{"model":"kimi-h3","prompt":"A cinematic slow orbit around a mountain lake at sunrise","duration":5,"resolution":"768p","aspect_ratio":"16:9"}'
# poll until terminal
curl "https://api.artany.ai/v1/jobs/{job_id}" \
-H "Authorization: Bearer $ARTANY_API_KEY"