About InfiniteTalk
InfiniteTalk is an open-source model from MeiGen-AI, released in August 2025 under the Apache-2.0 license. It turns one photo and a voice recording into a video of that person speaking, at 480p or 720p, for recordings from 3 seconds to 10 minutes. In a group photo, a mask marks which person should talk.
Call it with the model name meigen-ai/infinitetalk-v1/lip-sync on POST /jobs/createTask, from $0.039.
Introduction
Redubbing that changes the delivery
MeiGen-AI’s InfiniteTalk is a standard open-weight model under Apache-2.0. The meigen-ai/infinitetalk-v1/lip-sync endpoint accepts footage plus a new recording and produces an MP4. Mouth animation, head movement, facial expression, and posture can be regenerated together. Choose this workflow when a new line should influence the visible performance.
MeiGen-AI describes the method as sparse-frame video dubbing: the source supplies identity, background, and camera context while audio guides the generated movement. Review the complete performance rather than expecting the source motion to be reproduced frame for frame. The voice comes from your uploaded recording; the endpoint does not clone a voice.
Six request fields
| Field | Required | Accepted value and role |
|---|---|---|
video_url | Yes | Source footage with the intended person clearly visible. |
audio_url | Yes | A recording lasting 3–600 seconds; supplies the voice and output timing. |
mask_url | No | A speaker-selection image; mark the intended person white when several people appear. |
prompt | No | Up to 2,000 characters describing expression, action, or lighting. |
resolution | No | 480p by default, or 720p at twice the per-second rate. |
seed | No | A fixed random seed for comparing a single changed input; repeatability is not guaranteed. |
There is no separate duration field: prepare the recording to the length you want. A source clip longer than the recording is shortened in the result. The florist example also returned 6.10 seconds from 6.1 seconds of audio paired with 6.04 seconds of footage; this short extension does not establish behavior for every difference in input length.
Count audio seconds and choose a tier
The 720p rate is twice the 480p rate. Source-video duration, speaker masking, prompt wording, and seed selection do not change that rule.
For the florist request, the 6.1-second recording counted as seven seconds at the 720p tier. Start by checking the entire recording, including pauses and its ending, because that track determines both the speaking length and chargeable duration.
Make the speaker visible
Pick footage where you can clearly assess the intended person’s expression and movement. For a group scene, supply mask_url with white over the intended speaker. The mask selects the person; it is not a replacement for the audio track.
Keep instructions about the performance
Put the actual dialogue in the recording. A useful optional direction for a quiet workshop scene is: “The speaker looks up with a small surprised smile and a gentle shoulder movement; keep the warm workshop light steady.” Adapt the direction to the scene supplied in your footage. The text field can guide visible delivery but cannot replace audio_url.
Compare a single changed instruction while retaining the same media, seed, and resolution. A fixed seed helps make that comparison useful; it does not promise identical frames or a repeated gesture. Check what changed by watching both outputs with sound.
Observed video results
Our rainy-florist-surprise example began with a calm 6.04-second workshop clip at 864×496. The 6.1-second recording moves from hesitation to delight. With a motion prompt and resolution: 720p, the result measured 1264×720 at 25 fps and lasted 6.10 seconds. Her head lifted, her expression brightened, and her shoulders and upper body moved as her mouth followed the words.
In a separate 480p test, a 5.4-second source with a 4.55-second recording returned 4.55 seconds at 512×768 and 25 fps. The mouth and head posture followed the replacement line. These are measurements of short clips, not a guarantee of a particular frame size or sustained quality across a ten-minute recording.
Retrieve, review, and choose a workflow
Requests run asynchronously. Keep the task ID after submission, then poll for completion or use a callback to collect the MP4. Review speech timing, expression changes, head movement, and the scene through the end of the recording.
Use the photo-led talking-avatar endpoint in InfiniteTalk when your source is a still image. For existing footage, compare Sync Lipsync 2 if preserving the source speaking style and choosing how mismatched durations are handled matter to the edit. Consider OmniHuman 1.5 for a portrait workflow with text direction for gestures, expression, and camera movement. Select according to the input material and controls you need, then assess a short result with your own authorized media.
Request parameters
| Parameter | Required | Description |
|---|---|---|
audio_url | Yes | The voice or music track to use. |
mask_url | No | Paint over the area that should change. |
prompt | No | Describe what you want to make. |
resolution | No | Choose an available output size. 480p · 720p · Default: 480p |
seed | No | Use the same seed and settings for more controlled comparisons. ≥ 0 · ≤ 2147483647 |
video_url | Yes | The clip to work on. |
Pricing
Prices are in USD and charged to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.
| Option | USD per second |
|---|---|
| resolution 480p | $0.039 |
| resolution 720p | $0.078 |
Send the same request body to POST /jobs/quote to see the estimated and maximum USD charge before running. A quote does not run the task or charge your balance.
API quick start
Call the InfiniteTalk API in three steps
- Create an API key. Sign in to PopNow and create a key in the API console. Keys start with sk-popnow-. API calls are billed in USD from your PopNow balance.
- Send the request. POST /jobs/createTask with model meigen-ai/infinitetalk-v1/lip-sync and an input object that follows the parameter table. The response returns a task ID straight away.
- Get the result. Poll GET /jobs/recordInfo with the task ID, or pass a callBackUrl and we post the result to it. Download the output links when the task succeeds.
# 1. Create the task. Keep the Idempotency-Key if you retry after a network error.
export IDEMPOTENCY_KEY="$(uuidgen)"
curl -X POST https://api.popnow.ai/api/v1/jobs/createTask \
-H "Authorization: Bearer $POPNOW_API_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: $IDEMPOTENCY_KEY" \
-d '{
"model": "meigen-ai/infinitetalk-v1/lip-sync",
"input": {
"video_url": "https://media.popnow.ai/effects/flower-bloom-1abd6efb.mp4",
"audio_url": "https://media.popnow.ai/shortfilm/voices/Aoede-46e4c427.mp3",
"resolution": "480p"
}
}'
# 2. Copy data.taskId from the response, then poll until state is succeeded, failed, canceled or expired.
TASK_ID="TASK_ID_FROM_RESPONSE"
curl "https://api.popnow.ai/api/v1/jobs/recordInfo?taskId=$TASK_ID" \
-H "Authorization: Bearer $POPNOW_API_KEY"Replace the prompt and any links with your own. The API tab has the response, polling and callback reference.
InfiniteTalk API questions
How much does the InfiniteTalk API cost?
InfiniteTalk starts at $0.039 per second with Lip Sync, its lowest-priced task. Prices by option: 480p $0.039 per second, 720p $0.078 per second. API usage is charged in USD to your PopNow balance. When a task is accepted, we reserve the maximum quoted amount from your API balance. On completion, we charge for actual usage, never more than the hold, and release the remainder. If the task fails, the hold is released.
What are the limits of the InfiniteTalk · Lip Sync API?
Resolutions: 480p · 720p; Prompt length (characters): 2,000; Largest upload: 90 MB. Requests outside these limits are rejected before anything is charged.
More about InfiniteTalk
- Endpoints
- InfiniteTalk · AI Avatar