LTX-2 Retake performs targeted retakes on any section of a video—replace visuals, audio, or both—while preserving timing and continuity with $0.1 per output video second. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.
Idle
$0.1per run·~10 / $1
Change the time in the video from daytime to nighttime.
From 2s to 8s. Make the robot, saying in English: 'Master, the people outside are dangerous.' Keep the background, lighting, and camera angle the same, with natural robotic lip-sync and a worried tone.
Remove the man on the left.
Replace the man in the video with a woman.
Change the woman's clothes in the video to red. Do not alter the camera, composition, background, lighting, faces, or any other details.
LTX-2 Retake is Lightricks’ next-generation AI video refinement model for “reshooting” part of a finished clip—without bringing the crew back or regenerating the whole video.
Instead of rebuilding the scene from scratch, Retake lets you describe what should change in natural language while it preserves the rest of the footage, keeping lighting, motion continuity, composition, and environment consistent.
Because high-quality revision shouldn’t require starting from scratch. LTX-2 Retake gives you surgical control over moments inside an existing video—fast, flexible, and production-ready.
| Parameter | Description |
|---|---|
| prompt* | Natural-language instructions for the retake: what to change, how it should look or sound, and (optionally) which time span in the clip you want to modify. |
| video* | Original clip to refine. |
| mode | Editing mode: replace_video (change visuals, keep audio), replace_audio (keep visuals, regenerate audio), or replace_audio_and_video (update both picture and sound). |
Output resolution is currently fixed at 1080p in this integration.
👉 Time ranges: If you need to target a specific segment, write the time window directly into the prompt, for example: “From 00:05 to 00:10, change the time of day from afternoon to nighttime while keeping the same actor and camera angle.”
| Actual clip duration | Billed seconds | Effective rate per second | Total price |
|---|---|---|---|
| ≤ 2 s | 2 s | $0.10 / s | $0.20 |
| 5 s | 5 s | $0.10 / s | $0.50 |
| 10 s | 10 s | $0.10 / s | $1.00 |
| ≥ 16 s | 16 s (capped) | $0.10 / s | $1.60 |
Grab a WaveSpeedAI API key, then call POST https://api.wavespeed.ai/api/v3/lightricks/ltx-2-retake with your input as JSON. The endpoint returns a prediction id. Start polling the result endpoint around every 2 seconds, increase the interval for long-running tasks, and stop on any terminal status. On completed, read output values from data.outputs. Examples for Ltx 2 Retake below.
set -euo pipefail
: "${WAVESPEED_API_KEY:?Set WAVESPEED_API_KEY}"
REQUEST_BODY=$(cat <<'JSON'
{
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
"mode": "replace_audio_and_video"
}
JSON
)
# 1. Submit the prediction.
SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
-X POST "https://api.wavespeed.ai/api/v3/lightricks/ltx-2-retake" \
-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/lightricks/ltx-2-retake";
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 ${apiKey}`,
"Content-Type": "application/json",
},
body: JSON.stringify({
"prompt": "A cinematic shot of a city at sunset, soft golden light",
"video": "https://interactive-examples.mdn.mozilla.net/media/cc0-videos/flower.mp4",
"mode": "replace_audio_and_video"
}),
});
const task = body.data ?? body;
if (!task.id) throw new Error("Submission response did not contain a prediction id");
const resultUrl = `https://api.wavespeed.ai/api/v3/predictions/${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",
"mode": "replace_audio_and_video"
}
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/lightricks/ltx-2-retake", 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)Ltx 2 Retake is a Lightricks model for video editing, exposed as a REST API on WaveSpeedAI. LTX-2 Retake performs targeted retakes on any section of a video—replace visuals, audio, or both—while preserving timing and continuity with $0.1 per output video second. 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/lightricks/lightricks-ltx-2-retake.
Ltx 2 Retake starts at $0.10 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`, `mode`. 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/lightricks/lightricks-ltx-2-retake.
Median end-to-end generation time on WaveSpeedAI is around 74 seconds per request, based on recent successful runs. Queue time varies with global demand; live status is visible in the prediction record.
Commercial usage rights depend on the model's license, set by its provider (Lightricks). The license summary appears on the model card above; see WaveSpeedAI's Terms of Service for platform-level conditions.