Minimax H3 Image To Video Spicy API Documentation
Playground
Try it on WaveSpeedAI!MiniMax H3 Open Weights Spicy Image to Video is a fast AI image-to-video generation model that creates high-quality cinematic clips from images, optimized for scalable content generation with smooth, expressive animations, stable aesthetics, and native stereo audio. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Features
MiniMax H3 Image-to-Video Spicy Open Weights animates a start image into a video with prompt-guided motion. The output canvas follows the input image’s aspect ratio, and you can optionally provide a last_image to guide the ending frame.
Why Choose This?
-
Image-to-video generation
Animate a start image into a short video. -
Optional prompt control
Add a prompt to describe the scene, action, camera movement, and mood. -
Start-and-end-frame guidance
Useimageas the first frame and optionally providelast_imageto guide the final frame. -
Native canvas option
Use768pfor the model’s native canvas,480pfor faster, lower-cost generation,540pfor a mid tier at 1.5x the 480p price, or1080pfor full-HD output at 2x the 768p price. -
Flexible duration
Choose an output duration from3to15seconds. -
Seed support
Set a seed for more reproducible generation, or use a negative value for random generation.
Parameters
| Parameter | Required | Description |
|---|---|---|
| image | Yes | Start image URL to guide the video generation. The output canvas follows this image’s aspect ratio. |
| prompt | No | Describe the scene, action, camera movement, and mood for the video. |
| last_image | No | Optional last-frame image URL. When provided, the video interpolates from the first frame to this frame. |
| resolution | No | Output video resolution: 480p, 540p, 768p, or 1080p. 768p is the model’s native canvas; 480p is a faster, lower-cost tier; 540p is a mid tier at 1.5x the 480p price; 1080p is the highest-quality full-HD tier. Default: 768p. |
| duration | No | Output video duration in seconds. Options: 3 to 15. Default: 5. |
| seed | No | Random seed for generation. A negative value means a random seed will be used. |
How to Use
- Upload a start image — Provide the image that should define the first frame and canvas aspect ratio.
- Write a prompt optional — Describe the action, motion, camera movement, mood, and scene direction.
- Add a last image optional — Use
last_imagewhen you want to guide the final frame. - Choose resolution — Use
480pfor faster generation,540pfor a mid-price step up,768pfor the native canvas, or1080pfor full-HD output. - Set duration — Choose a duration from
3to15seconds. - Set seed optional — Use a fixed seed for reproducible results, or a negative value for random generation.
- Submit — Generate the final image-to-video output.
Pricing
Pricing is based on generated video duration and selected resolution. The price is charged per second of video, and the per-second rate depends on the resolution.
| Resolution | Per second (duration 5 s) | duration 5 s | duration 15 s |
|---|---|---|---|
| 480p | |||
| 540p | |||
| 768p | |||
| 1080p |
Duration is capped at 15 seconds for pricing.
Example Costs
| Resolution | duration 3 s | duration 5 s | duration 10 s | duration 15 s |
|---|---|---|---|---|
| 480p | ||||
| 540p | ||||
| 768p | ||||
| 1080p |
With default settings (768p, duration 5 s), one run costs $0.40 $0.20 per run.
prompt, last_image, and seed do not add separate charges.
Best Use Cases
- Image animation — Turn a still image into a short motion clip.
- Character motion — Animate portraits, characters, or illustrated subjects.
- Start-and-end-frame transitions — Use
imageandlast_imageto guide both ends of the video. - Creative prototyping — Test motion, camera direction, and scene mood from a single image.
- Social video content — Generate short clips from static images for vertical, square, or landscape-style compositions.
- Prompt variation testing — Compare different actions or camera movements from the same start image.
Pro Tips
- Use a clear, high-quality start image for stronger subject and composition preservation.
- Add
last_imagewhen the final pose, framing, or ending state matters. - Describe visible action clearly, such as turning, walking, looking up, zooming in, or moving through the scene.
- Include camera movement and mood when the shot direction matters.
- Use
480pfor quick testing,540pfor a step up, and768por1080pfor higher-resolution output. - Use a fixed
seedwhen comparing prompt or parameter changes.
Related Models
- MiniMax H3 Text-to-Video Open Weights — Generate video with native stereo audio directly from a text prompt.
- MiniMax H3 Image-to-Video Open Weights — Animate a first-frame image with prompt-guided motion and audio.
- MiniMax H3 Reference-to-Video Open Weights — Generate video from image, video, or audio references with prompt guidance.
Authentication
For authentication details, please refer to the Authentication Guide.
API Endpoints
Submit Task & Query Result
set -euo pipefail
export WAVESPEED_API_KEY="your-api-key"
REQUEST_BODY=$(cat <<'JSON'
{
"image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
"resolution": "768p",
"duration": 5
}
JSON
)
# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
--connect-timeout 10 --max-time 60 \
-X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/image-to-video-spicy" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
-H "Content-Type: application/json" \
-d "${REQUEST_BODY}"); then
printf 'Submit failed. %s\n' "${CHECK}" >&2
exit 1
fi
TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"
# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
# Transient errors (429, 5xx) are retried, honoring Retry-After.
HTTP_CODE=$(curl --silent --show-error --fail-with-body \
--connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
--output "${BODY_FILE}" --write-out '%{http_code}' \
"${RESULT_URL}" \
-H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
case "${HTTP_CODE}" in
2??) ;;
000|429|5??) # dropped connection or temporary error: keep polling
printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
sleep 10; continue ;;
*) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
cat "${BODY_FILE}" >&2; exit 1 ;;
esac
RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')
case "${STATUS}" in
completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted)
printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneRetry temporary errors only while polling, never the submit request. Errors, retries and timeouts
Parameters
Task Submission Parameters
Request Parameters
| Parameter | Type | Required | Default | Range | Description |
|---|---|---|---|---|---|
| image | string | Yes | - | Start image URL to guide the video generation. The output canvas follows this image's aspect ratio. | |
| prompt | string | No | - | Describe the scene, action, camera movement, and mood for the video. | |
| last_image | string | No | - | - | Optional last-frame image URL. When provided, the video interpolates from the first frame to this frame. |
| resolution | string | No | 768p | 480p, 540p, 768p, 1080p | Output video resolution. 768p is the model's native canvas; 480p is a faster, lower-cost tier; 540p is a mid tier at 1.5x the 480p price. 1080p is the highest-quality full-HD tier at 2x the 768p price (generation takes longer). |
| duration | integer | No | 5 | 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15 | Output video duration in seconds. |
| seed | integer | No | - | - | The random seed to use for the generation. A negative value means a random seed will be used. |
Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.
Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data.id | string | Unique identifier for the prediction, Task Id |
| data.model | string | Model ID used for the prediction |
| data.outputs | array | Output values, usually URL strings; some models return text strings or structured result objects (empty when status is not completed) |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Task status. completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses. |
| data.created_at | string | ISO timestamp of when the request was created (e.g., “2023-04-01T12:34:56.789Z”) |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |
Result Request Parameters
| Parameter | Type | Required | Default | Description |
|---|---|---|---|---|
| id | string | Yes | - | Task ID |
Result Response Parameters
| Parameter | Type | Description |
|---|---|---|
| code | integer | HTTP status code (e.g., 200 for success) |
| message | string | Status message (e.g., “success”) |
| data | object | The prediction data object containing all details |
| data.id | string | Unique identifier for the prediction |
| data.model | string | Model ID used for the prediction |
| data.outputs | array<string | object> | Array of generated outputs (empty when status is not completed). Items are usually URL strings, but may be text strings or structured result objects, depending on the model. |
| data.urls | object | Object containing related API endpoints |
| data.status | string | Status: completed is successful; failed, cancelled, timeout, and deleted are failure terminal statuses |
| data.created_at | string | ISO timestamp of when the request was created |
| data.error | string | Error message (empty if no error occurred) |
| data.timings | object | Object containing timing details |
| data.timings.inference | integer | Inference time in milliseconds |