MiniMax H3 Open Weights Video-Extend appends a new cinematic continuation to an existing video. A fresh segment with native stereo audio is generated from the input video's last frame and a natural-language prompt, then the original and new segment are concatenated into a single output. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Idle
$0.1per run·~10 / $1
The camera keeps gliding forward past the lighthouse and out over the open sea as the sun sinks toward the horizon, seabirds sweeping across the frame, waves rolling gold and amber below, the light beam flickering on, wind and surf continuing seamlessly.
Continue seamlessly from the final moment. The man remains facing the giant movie screen for a second, then slowly turns around. Behind him, an old red convertible is now parked beside his car, although the road was empty moments ago. A young woman in a pale vintage dress stands beside the convertible, watching him silently. He looks back toward the movie screen. The black-and-white film now shows both of them standing together, but the woman on the screen suddenly begins walking away. The real woman turns in the same direction and slowly starts walking toward the abandoned cinema building. After hesitating, the man follows her. The camera continues the same slow cinematic movement, shifting from the man’s reaction into a wider two-character shot, then follows them from behind toward the dark cinema building. Preserve the same man, clothing, desert location, vehicles, twilight lighting, film-screen appearance, camera style, and visual continuity. Slow suspense, natural body movement, cinematic mystery, no scene reset.
MiniMax H3 Video Extend Open Weights extends an existing video by generating a new segment from the input video's last frame. The new segment is appended to the original video, with prompt-guided motion, scene continuation, and native audio generation.
Video extension workflow
Continue an existing video with a newly generated segment.
Last-frame continuation
Generate the new segment from the final frame of the input video for smoother visual continuity.
Prompt-guided motion
Describe the continuation, including action, camera movement, scene development, and soundtrack direction.
Optional last-frame target
Use last_image to guide where the extended segment should end.
Native audio generation
Generate audio together with the new video segment.
Flexible resolution options
Choose 480p for faster, lower-cost generation or 768p for the model's native canvas.
| Parameter | Required | Description |
|---|---|---|
| prompt | Yes | Text description of the continuation, including motion, scene development, camera movement, and soundtrack direction. Audio is generated natively together with the video. |
| video | Yes | URL of the video to extend. The new segment is generated from its last frame and appended to the original video. |
| last_image | No | Optional target last-frame image URL. When provided, the new segment interpolates from the input video's last frame to this image. |
| resolution | No | Resolution of the new segment: 480p or 768p. 768p is the model's native canvas; 480p is a faster, lower-cost tier. Default: 480p. |
| duration | No | Length in seconds of the new segment to append. Options: 3 to 15. Default: 5. |
| seed | No | Random seed for generation. A negative value means a random seed will be used. |
last_image when you want to guide the final frame of the new segment.480p for faster generation or 768p for the native canvas.3 to 15 seconds.Pricing is based on the generated extension duration and selected resolution.
| Resolution | Per second | 5s | 15s |
|---|---|---|---|
| 480p | $0.04 | $0.20 | $0.60 |
| 768p | $0.08 | $0.40 | $1.20 |
Duration is capped at 15 seconds for pricing.
| Resolution | 3s | 5s | 10s | 15s |
|---|---|---|---|---|
| 480p | $0.12 | $0.20 | $0.40 | $0.60 |
| 768p | $0.24 | $0.40 | $0.80 | $1.20 |
video, last_image, prompt, and seed do not add separate charges.
last_image when the extension needs to end on a specific visual state.last_image when the final pose, framing, or ending state matters.480p for quick testing and 768p for higher-resolution output.seed when comparing prompt or parameter changes.prompt and video are required.duration supports values from 3 to 15 seconds.last_image is optional and guides the target ending frame of the new segment.Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/video-extend with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read output values from data.outputs. Examples for Minimax H3 Video Extend below.
set -euo pipefail
: "${WAVESPEED_API_KEY:?Set WAVESPEED_API_KEY}"
REQUEST_BODY=$(cat <<'JSON'
{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
"resolution": "480p",
"duration": 5
}
JSON
)
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/video-extend" \
-H "Content-Type: application/json" \
-H "Authorization: Bearer $WAVESPEED_API_KEY" \
-d "$REQUEST_BODY")
TASK=$(printf '%s' "$SUBMIT_RESPONSE" | jq 'if has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "$TASK" | jq -r '.id')
if [ -z "$PREDICTION_ID" ] || [ "$PREDICTION_ID" = "null" ]; then
printf 'Submission response did not contain a prediction id
' >&2
exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/$PREDICTION_ID/result"
# 2. Poll until the prediction finishes.
while true; do
RESPONSE=$(curl --silent --show-error --fail-with-body "$RESULT_URL" \
-H "Authorization: Bearer $WAVESPEED_API_KEY")
RESULT=$(printf '%s' "$RESPONSE" | jq 'if has("data") then .data else . end')
STATUS=$(printf '%s' "$RESULT" | jq -r '.status')
case "$STATUS" in
completed) printf '%s\n' "$RESULT" | jq '.outputs'; break ;;
failed|cancelled|timeout|deleted) printf '%s\n' "$RESULT" | jq . >&2; exit 1 ;;
*) sleep 2 ;;
esac
doneconst submitUrl = "https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/video-extend";
const apiKey = process.env.WAVESPEED_API_KEY;
if (!apiKey) throw new Error('Set WAVESPEED_API_KEY');
async function requestJson(url, options = {}) {
const response = await fetch(url, options);
if (!response.ok) throw new Error(await response.text());
return response.json();
}
// 1. Submit the prediction.
const body = await requestJson(submitUrl, {
method: "POST",
headers: {
"Authorization": `Bearer ${apiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
"resolution": "480p",
"duration": 5
}),
});
const task = body.data ?? body;
if (!task.id) throw new Error("Submission response did not contain a prediction id");
const resultUrl = `https://api.wavespeed.ai/api/v3/predictions/${task.id}/result`;
// 2. Poll until the prediction finishes.
while (true) {
const resultBody = await requestJson(resultUrl, {
headers: { "Authorization": `Bearer ${apiKey}` },
});
const result = resultBody.data ?? resultBody;
if (result.status === "completed") {
console.log(result.outputs);
break;
}
if (["failed", "cancelled", "timeout", "deleted"].includes(result.status)) throw new Error(JSON.stringify(result));
await new Promise(resolve => setTimeout(resolve, 2000));
}import json
import os
import time
from urllib.request import Request, urlopen
api_key = os.environ["WAVESPEED_API_KEY"]
headers = {"Authorization": f"Bearer {api_key}", "Content-Type": "application/json"}
payload = {
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
"resolution": "480p",
"duration": 5
}
def request_json(url, data=None):
request = Request(url, data=data, headers=headers, method="POST" if data else "GET")
with urlopen(request) as response:
return json.load(response)
# 1. Submit the prediction.
body = request_json("https://api.wavespeed.ai/api/v3/wavespeed-ai/minimax-h3/video-extend", json.dumps(payload).encode())
task = body.get("data", body)
if not task.get("id"):
raise RuntimeError("Submission response did not contain a prediction id")
result_url = f"https://api.wavespeed.ai/api/v3/predictions/{task['id']}/result"
# 2. Poll until the prediction finishes.
while True:
result_body = request_json(result_url)
result = result_body.get("data", result_body)
status = result.get("status")
if status == "completed":
print(result.get("outputs", []))
break
if status in {"failed", "cancelled", "timeout", "deleted"}:
raise RuntimeError(result)
time.sleep(2)Minimax H3 Video Extend is a WaveSpeedAI model for video extension, exposed as a REST API on WaveSpeedAI. MiniMax H3 Open Weights Video-Extend appends a new cinematic continuation to an existing video. A fresh segment with native stereo audio is generated from the input video's last frame and a natural-language prompt, then the original and new segment are concatenated into a single output. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing. You can call it programmatically or try it from the playground above.
POST your input parameters to the model's REST endpoint (shown in the API tab of this playground) with your WaveSpeedAI API key in the Authorization header. Submission returns a prediction ID. Poll the result endpoint starting around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. The playground generates production-oriented Python, JavaScript, and cURL examples with timeouts, transient-error handling, and safe GET retries. Full request/response shape is documented at https://wavespeed.ai/docs/docs-api/wavespeed-ai/minimax-h3-video-extend.
Minimax H3 Video Extend starts at $0.20 per run. That figure is the base price — the final charge scales with the parameters you set in the form (output size, length, count, references, or whatever knobs this model exposes), so a higher-quality or larger output costs more than a minimal one. The exact cost for your current input is shown live next to the Generate button before you submit, and the actual per-call charge is recorded on the prediction afterwards.
Key inputs: `prompt`, `video`, `resolution`, `duration`, `seed`, `last_image`. The full JSON schema (types, defaults, allowed values) is rendered above the Generate button and mirrored in the API reference at https://wavespeed.ai/docs/docs-api/wavespeed-ai/minimax-h3-video-extend.
Sign up for a free WaveSpeedAI account to claim starter credits, copy your API key from /accesskey, then call the endpoint shown in the API tab of the playground. The playground also auto-generates a code sample in Python, JavaScript, or cURL for the parameters you've set.
Commercial usage rights depend on the model's license, set by its provider (WaveSpeedAI). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.