seedinteger
Seed for the continuation. Random when omitted.
minimax/h3-max-turbo/video-extendMiniMax H3 Max Turbo, the fastest tier of the H3 family post-trained by fal, continues an existing video with 1-15 seconds of new footage with native audio.
Try the model
Customize your input with more control.
What would you like to do next?
Select an example to load its inputs and result into the playground.
{
"duration": 1,
"output": "continuation",
"prompt": "the red panda keeps walking",
"resolution": "480P",
"video": "https://static.sandbase.ai/examples/b2271e8d-deb2-4284-9c26-65e71970825c/input/01a2e4dc-561e-48b4-8e31-86c1a102f2cd.mp4"
}MiniMax H3 Max Turbo video extend: continue an existing video with 1–15 seconds of new footage; return the joined clip or only the continuation.
| Field | Type | Default | Notes |
|---|---|---|---|
prompt | string, required | — | Up to 50,000 characters. |
aspect_ratio | 21:9 / 16:9 / 4:3 / 1:1 / 3:4 / 9:16 | — | Output frame shape. |
video | video URL, required | — | Clip to continue. |
resolution | 480P / 768P / 1080P / 2K | 768P | Sets the per-second rate. |
duration | integer 1–15 | 5 | Seconds of new footage. |
seed | integer | random | Fix it to compare prompt changes. |
reference_audio_urls | audio URL[] | — | Optional reference audio for the new footage. |
output | extended / continuation | extended | extended returns source + new footage; continuation returns only the new part. |
You pay for new seconds: $0.025 at 480P, $0.04 at 768P, $0.08 at 1080P, $0.16 at 2K.
| Output | 480P | 768P | 1080P | 2K |
|---|---|---|---|---|
| 5s (default) | $0.125 | $0.20 | $0.40 | $0.80 |
| 10s | $0.25 | $0.40 | $0.80 | $1.60 |
| 15s | $0.375 | $0.60 | $1.20 | $2.40 |
Failed runs are not charged.
Other H3 Max Turbo endpoints: minimax/h3-max-turbo/text-to-video, minimax/h3-max-turbo/image-to-video, minimax/h3-max/text-to-video (H3 Max).
Start building
Submit this model through the SandBase Unified Run API.
# The quoted heredoc keeps Unicode and shell metacharacters unchanged.
result=$(curl --fail-with-body --silent \
-X POST "https://api.sandbase.ai/v1/run" \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
-H "Content-Type: application/json" \
--data-binary @- <<'SANDBASE_JSON'
{
"model": "minimax/h3-max-turbo/video-extend",
"video": "https://storage.googleapis.com/falserverless/example_inputs/flux-3-red-panda.mp4",
"output": "extended",
"prompt": "The red panda keeps walking along the log, then looks up at a bird.",
"duration": 5,
"resolution": "768P"
}
SANDBASE_JSON
)
run_id=$(printf '%s' "$result" | jq -r .id)
for attempt in $(seq 1 120); do
status=$(printf '%s' "$result" | jq -r .status)
case "$status" in completed|failed|timeout) break ;; esac
sleep 2
result=$(curl --fail-with-body --silent \
-H "Authorization: Bearer $SANDBASE_API_KEY" \
"https://api.sandbase.ai/v1/run/$run_id")
done
status=$(printf '%s' "$result" | jq -r .status)
[ "$status" = completed ] || { echo "Generation ended: $status" >&2; exit 1; }
printf '%s\n' "$result"Create a SandBase API key and set it in your server environment. All requests use Bearer authentication.
export SANDBASE_API_KEY="YOUR_API_KEY"Keep the key on your server. Do not expose it in browser code or commit it to a repository.
POST to /v1/run with this model ID and its inputs. The request example above includes submission and polling.
Use the returned run ID to poll until the status is completed, failed or timeout. Read output only after completed.
curl "https://api.sandbase.ai/v1/run/$RUN_ID" \
-H "Authorization: Bearer $SANDBASE_API_KEY"Pass a publicly accessible file URL in the model’s image, video or audio field. Supported formats and limits are shown beside each Playground input.
Use the upload or replace control on supported image fields in the Playground. Video URL fields accept a hosted driving video; they do not imply video-upload support.
seedintegerSeed for the continuation. Random when omitted.
videostringrequiredURL of the video to extend. Must be 1.625 s to 60 s long, up to 50 MB, aspect ratio between 0.4 and 2.5.
outputstring'extended' returns the whole source video followed by the new footage. 'continuation' returns only the generated continuation.
Default: "extended"
Possible values: "extended", "continuation"
promptstringrequiredWhat happens next. Describe the continuation, not the source.
durationintegerSeconds of new footage to add after the source. The actual duration may be slightly longer.
Default: 5
Minimum: 1
Maximum: 15
resolutionstringOutput video resolution.
Default: "768P"
Possible values: "480P", "768P", "1080P", "2K"
aspect_ratiostringOutput aspect ratio. Matches the source video when omitted.
Possible values: "21:9", "16:9", "4:3", "1:1", "3:4", "9:16"
reference_audio_urlsarrayOptional audio references for the continuation (2–15 seconds each, 15 seconds combined). The full extended video retains its original source soundtrack, and its ending still conditions the transition. These clips guide the newly generated audio instead of using that soundtrack as an audio reference. Assign each clip a role in the prompt, such as 'Audio 1 for the voice' and 'Audio 2 for background music'.
{
"model": "minimax/h3-max-turbo/video-extend",
"video": "https://storage.googleapis.com/falserverless/example_inputs/flux-3-red-panda.mp4",
"output": "extended",
"prompt": "The red panda keeps walking along the log, then looks up at a bird.",
"duration": 5,
"resolution": "768P"
}A completed run exposes its result in output. The following is the model’s saved example, not a new generation.
[
{
"poster": "https://static.sandbase.ai/examples/posters/95b7be92f3f57c6dcfbf.jpg",
"source": "Self organization completed prediction",
"source_url": "https://www.sandbase.ai",
"url": "https://static.sandbase.ai/examples/b2271e8d-deb2-4284-9c26-65e71970825c/output/af9f4df3-3e54-4a73-83b1-6ce4adb683c1.mp4"
}
]