MCP / Models / Reference video, editing and lip sync

LTX 2.5 Fast Audio to Video MCP: quick video timed to audio

LTX 2.5 Fast makes a video timed to your audio clip of 2 to 20 seconds, from a prompt and an optional first frame. Run it from Claude Code, Codex, Cursor or any MCP client through one FlyMy.AI MCP, or call its API.

No API key$0.13 / sAudio to videoLightricksCommercial use
LTX 2.5 Fast Audio to Video sample from the FlyMy.AI catalog

$ YOUR NEXT RUN

Claude CodeCodexCursorAny agent

Use LTX 2.5 Fast Audio to Video in your assistant

Paste this prompt into Claude Code, Codex, Cursor or any agent. It reads the guide and helps you connect.

Read https://flymy.ai/skill.md and https://flymy.ai/mcp/models/ltx-2-5-fast-audio-to-video.md. Help me connect the FlyMy.AI MCP to this assistant, then run LTX 2.5 Fast Audio to Video (flymyai/ltx-2_5-fast-audio-to-video) for this task: Make a video timed to the audio clip I attach with LTX 2.5 Fast Audio to Video: neon city streets at night, the camera drifting forward.
Open FlyMy.AIDownload brief

What LTX 2.5 Fast Audio to Video does

Send audio, a WAV, MP3 or M4A file of 2 to 20 seconds (music, dialogue or a voice-over), and prompt, required, up to 5,000 characters, describing the scene, the action and the camera; with image, an optional first frame, it says how the picture is animated. aspect_ratio is auto, which follows the picture (16:9 without one), 16:9 or 9:16. Output: video, one MP4 clip timed to the audio. Use it for music-driven content, dialogue-led shorts and ads keyed to a track; for final, high-fidelity output use LTX 2.5 Pro Audio to Video.

It is billed by the seconds of the audio you send, rounded up, and the length of the audio is read before the run. It does not write music or speech (bring an audio file, or make one with a music or speech model), take a reference video or edit or extend a video.

Use only audio and pictures you have the right to use: a real person's voice or face needs that person's consent.

LTX 2.5 Fast Audio to Video inputs

Every input of flymyai/ltx-2_5-fast-audio-to-video as its live schema lists them. Returns video (array).

InputType, values and default
# audio
file required Allowed: audio file, as a URL or upload The audio the video is timed to: a WAV, MP3 or M4A file of 2 to 20 seconds, a link or an upload. Its length sets the price: $0.13 a second of audio, in whole seconds rounded up
# prompt
string required Describe the video: the scene, the action and the camera; with an image, how the picture is animated
# image
file optional Allowed: image file, as a URL or upload Optional first frame of the video: a link or an upload
# aspect_ratio
string optional Allowed: auto, 16:9, 9:16 Default: auto Shape of the video: 16:9 or 9:16, or auto to follow the picture (16:9 without one)

Ask your agent

Requests to paste as they are, once the FlyMy.AI MCP is connected.

“Make a video timed to the audio clip I attach with LTX 2.5 Fast Audio to Video: neon city streets at night, the camera drifting forward.”runs flymyai/ltx-2_5-fast-audio-to-video
“Turn my voice-over into a short dialogue scene with LTX 2.5 Fast Audio to Video, starting from the character picture I attach.”runs flymyai/ltx-2_5-fast-audio-to-video
“Cut a 9:16 music video to the track I attach with LTX 2.5 Fast Audio to Video, with a singer on a rooftop at dusk.”runs flymyai/ltx-2_5-fast-audio-to-video

LTX 2.5 Fast Audio to Video price

$0.13 per second of the audio you send, rounded up. A 10-second clip costs $1.30. A run at the default settings costs $1.30. A run costs $0.26 to $2.60, depending on audio. Billed to your FlyMy.AI balance; no seats, no subscription and no minimum.

SettingPrice
Any$0.13 per second

Call LTX 2.5 Fast Audio to Video from code

The same model, from an agent or from your own code. Keys stay in your FlyMy.AI account.

From an agent, through the MCP

The agent calls get_model_schema, then run_model with:

{
  "endpoint_id": "flymyai/ltx-2_5-fast-audio-to-video",
  "input": {
    "audio": "https://example.com/voice.mp3",
    "prompt": "A singer on a rooftop at dusk, the camera slowly circling her"
  }
}

From your code, over HTTP

Form fields, not JSON; files go as uploads or public URLs. The answer is a prediction_id; read the result at https://api.flymy.ai/api/v1/flymyai/ltx-2_5-fast-audio-to-video/predict/async/result/?request_id=<prediction_id>.

curl -X POST https://api.flymy.ai/api/v1/flymyai/ltx-2_5-fast-audio-to-video/predict/async/ \
  -H "X-API-KEY: $FLYMYAI_API_KEY" \
  -F "audio=@voice.mp3" \
  -F 'prompt=A singer on a rooftop at dusk, the camera slowly circling her'

Questions about LTX 2.5 Fast Audio to Video

Do I need a Lightricks API key for LTX 2.5 Fast Audio to Video?

No. LTX 2.5 Fast Audio to Video runs on FlyMy.AI's account, so you need no Lightricks key or plan: sign in to FlyMy.AI once and each run is billed to your FlyMy.AI balance.

How much does LTX 2.5 Fast Audio to Video cost?

$0.13 per second of the audio you send, rounded up. A 10-second clip costs $1.30. There is no subscription or minimum; the price comes from FlyMy.AI billing and was checked on 2026-11-18.

How do I use LTX 2.5 Fast Audio to Video in Claude Code, Codex or Cursor?

Add the FlyMy.AI MCP with one command and sign in, then ask in plain words, for example: "Make a video timed to the audio clip I attach with LTX 2.5 Fast Audio to Video: neon city streets at night, the camera drifting forward." Your agent reads the inputs with get_model_schema and runs flymyai/ltx-2_5-fast-audio-to-video with run_model.

What does LTX 2.5 Fast Audio to Video need as input?

LTX 2.5 Fast Audio to Video requires audio and prompt; it also takes 2 optional inputs, each listed on this page with its allowed values and default.

Can I call LTX 2.5 Fast Audio to Video from my own code?

Yes. Send form fields to https://api.flymy.ai/api/v1/flymyai/ltx-2_5-fast-audio-to-video/predict/async/ with your FlyMy.AI API key in the X-API-KEY header.

Go to the source

From the FlyMy.AI model catalog of 2026-11-18: inputs from the model's live schema, prices from FlyMy.AI billing. Requests are examples to paste, not recorded runs.