Gemini Omni Flash Video Edit API

google/gemini-omni-flash/video-edit

Gemini Omni Flash Video Edit revises one source video through natural-language instructions, with conversational targeting of what to change, native audio updates, and resolution from 720p to 4K. It applies the requested edits while preserving the retained subjects, camera continuity, and temporal flow you specify.

Input

Prompt

237/20000

Source video

Add source video

Required. Upload exactly one MP4, MOV, or WebM clip, up to 100 MiB.

Resolution

Aspect ratio

Output

Idle

Your generated video will appear here

Add your prompt and required media, review the settings, then click Run.

Continue with

Gemini Omni Flash Text to Video Gemini Omni Flash Image to Video Gemini Omni Flash Reference to Video

Examples

Remove the jogger from the video completely. Keep the fountain, path, trees, benches, sunlight, and the original locked camera unchanged. Seamless background inpainting, no new people or objects. No logos, no readable text, no watermark.

Transform the scene into winter: cover the meadow and the wooden fence with snow, overcast gray sky, gentle falling snowflakes. Keep the camera, fence position, and composition unchanged. No logos, no readable text, no watermark.

Change the time of day to golden dusk: warm orange sunlight, dramatic glowing clouds, the sun low on the horizon reflecting on the water. Keep the sailboat, its position, and the camera unchanged. No logos, no readable text, no watermark.

REST API

Quick Start

Authenticate with the API, submit the source video and edit prompt, then retrieve the revised clip using the task ID.

Connect to the Vidgo API

Create an API key, keep it only on your server, and send Authorization: Bearer VIDGO_API_KEY.

Endpoint: POSThttps://api.vidgo.ai/api/generate/submit
Authentication: Authorization: Bearer VIDGO_API_KEY

Submit one generation task

Fill in the parameters for this endpoint using the request example, then save the returned task_id to query progress and results.

Submit request

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "google/gemini-omni-flash/video-edit",
  "input": {
    "prompt": "Remove the jogger from the video completely. Keep the fountain, path, trees, benches, sunlight, and the original locked camera unchanged. Seamless background inpainting, no new people or objects. No logos, no readable text, no watermark.",
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "video_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-flash/video-edit/v1/01/input-source.mp4"
    ]
  }
}
JSON
)

RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

CODE=$(printf '%s' "$RESPONSE" | jq -r '.code // empty')
if [ "$CODE" != "0" ] && [ "$CODE" != "200" ]; then
  printf 'API error: %s
' "$RESPONSE" >&2
  exit 1
fi
printf '%s
' "$RESPONSE"

Wait for the result

Query with task_id, continue for not_started/running, and stop for finished/failed. On success, read data.files[].file_url.

Track status

GET https://api.vidgo.ai/api/generate/status/{task_id}

Poll status with a 2-second base interval, and increase the interval for longer tasks. Continue only while status is not_started or running, and stop once finished or failed. You can also specify callback_url in the request payload to receive webhook notifications.

not_startedrunningfinishedfailed

Submit response

{
  "code": 200,
  "data": {
    "task_id": "task-unified-...",
    "status": "running",
    "created_time": "2026-09-15T10:00:00Z"
  }
}

Finished response

{
  "code": 200,
  "data": {
    "task_id": "task-unified-...",
    "status": "finished",
    "files": [
      {
        "file_url": "https://storage.vidgo.ai/generated/video.mp4",
        "file_type": "video"
      }
    ],
    "created_time": "2026-08-22T10:00:00Z",
    "progress": 100,
    "error_message": null
  }
}

Complete runnable example

Expand for a complete script with HTTP and business-code checks, task_id validation, polling, terminal-state handling, and a timeout boundary.

Complete runnable example

set -euo pipefail
: "${VIDGO_API_KEY:?Set VIDGO_API_KEY in your environment}"

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "google/gemini-omni-flash/video-edit",
  "input": {
    "prompt": "Remove the jogger from the video completely. Keep the fountain, path, trees, benches, sunlight, and the original locked camera unchanged. Seamless background inpainting, no new people or objects. No logos, no readable text, no watermark.",
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "video_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-flash/video-edit/v1/01/input-source.mp4"
    ]
  }
}
JSON
)

SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --request POST \
  --url "https://api.vidgo.ai/api/generate/submit" \
  --header "Authorization: Bearer $VIDGO_API_KEY" \
  --header "Content-Type: application/json" \
  --data "$REQUEST_BODY")

TASK_ID=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.data.task_id // .task_id // empty')
BUSINESS_CODE=$(printf '%s' "$SUBMIT_RESPONSE" | jq -r '.code // empty')
if [ "$BUSINESS_CODE" != "0" ] && [ "$BUSINESS_CODE" != "200" ]; then
  printf 'Submit failed:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi
if [ -z "$TASK_ID" ]; then
  printf 'Submit response did not include task_id:
%s
' "$SUBMIT_RESPONSE" >&2
  exit 1
fi

START_TIME=$(date +%s)
POLL_DELAY=2
while true; do
  if [ $(( $(date +%s) - START_TIME )) -ge 600 ]; then
    printf 'Timed out after 600 seconds
' >&2
    exit 1
  fi
  STATUS_RESPONSE=$(curl --silent --show-error --fail-with-body \
    --url "https://api.vidgo.ai/api/generate/status/$TASK_ID" \
    --header "Authorization: Bearer $VIDGO_API_KEY")
  STATUS=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.data.status // .status // empty')
  BUSINESS_CODE=$(printf '%s' "$STATUS_RESPONSE" | jq -r '.code // empty')
  if [ "$BUSINESS_CODE" != "0" ] && [ "$BUSINESS_CODE" != "200" ]; then
    printf 'Status request failed:
%s
' "$STATUS_RESPONSE" >&2
    exit 1
  fi

  case "$STATUS" in
    finished)
      printf '%s' "$STATUS_RESPONSE" | jq -r '(.data.files // .files // [])[]?.file_url'
      break
      ;;
    failed)
      printf '%s' "$STATUS_RESPONSE" | jq -r '.data.error_message // .error_message // "Generation failed"' >&2
      exit 1
      ;;
    not_started|running)
      sleep "$POLL_DELAY"
      if [ "$POLL_DELAY" -lt 10 ]; then POLL_DELAY=$((POLL_DELAY + 1)); fi
      ;;
    *)
      printf 'Unexpected task status: %s
' "$STATUS" >&2
      exit 1
      ;;
  esac
done

Input parameters

The table lists available input parameters, types, and defaults. Request examples also include the required top-level model field.

Field	Type	Required	Default	Description
prompt	string	Yes	—	1–20,000 characters after trimming.
video_urls	string[]	Yes	—	Exactly one publicly accessible HTTP(S) video URL is required; missing, empty, or multiple entries are rejected. duration is not accepted; this endpoint imposes no source-video duration limit.
resolution	string	No	720p	720p, 1080p, or 4k.
aspect_ratio	string	No	16:9	16:9 or 9:16.

Response Fields

A successful submission returns a task ID. Status queries provide progress, output files, and error details when a task fails.

Field	Type	Description
code	integer	Application result code; successful responses use 0 or 200.
message	string	Human-readable message or error detail when present.
data.task_id	string	Task ID used in the status endpoint path.
data.status	string	not_started, running, finished, or failed.
data.created_time	string	Task creation time in date-time format.
data.progress	integer	Task progress from 0 to 100, when included in the response.
data.files[]	array	All output files from a successful task, in response order.
data.files[].file_url	string	Public URL for a generated video.
data.files[].file_type	string	File type, such as video.
data.error_message	string \| null	Failure detail when status is failed.

Task Lifecycle

Continue querying while the status is not_started or running. End polling at finished or failed, then process the output files or error details respectively.

not_started
The task was accepted and is waiting to begin.
running
Generation is in progress. Continue polling the same task_id.
finished
Generation succeeded. Read every video URL from data.files[].file_url.
failed
Generation stopped with an error. Read data.error_message and stop polling.

Polling and Errors

AuthenticationFor a 401 response, check the Bearer API key in Authorization, update the credentials, and retry.
ValidationFor a 400 response, use the response details to check required fields, parameter ranges, and available credits, then adjust and resubmit.
Network and timeoutIf a status query encounters a network error or timeout, retain the original task_id and retry the query, then handle the result according to the returned task status.
Polling intervalPoll status with a 2-second base interval, and gradually increase the interval for longer tasks.
Terminal statesContinue only for not_started or running. Stop immediately on finished or failed.
Callback optionProvide callback_url at the request top level to receive the final flat task object; polling remains available if delivery fails.

Endpoint limits

Specification	Value	Details
Input mode	Prompt + 1 source video	One public source clip plus a natural-language edit instruction.
Source video	MP4 / MOV / WebM	Exactly one public HTTP(S) video URL in video_urls.
Output	Video	Returns an asynchronous task ID; finished tasks include a revised video file with native audio.
Resolution	720p / 1080p / 4k	Default is 720p.
Aspect ratio	16:9 / 9:16	Default is 16:9.
Billing basis	Per generation	720p/1080p: 300 credits. 4k: 400 credits.

View the Gemini Omni Flash Video Edit API docs

Gemini Omni Flash Video Edit

Gemini Omni Flash Video Edit is Google DeepMind’s multimodal model for conversational video revision. Upload one public source video URL, describe what to change and what to keep, then generate an updated clip with synchronized audio. Choose 720p, 1080p, or 4K with 16:9 or 9:16 framing—ideal for ad varianting, dialogue retakes, wardrobe or object swaps, and soundtrack refreshes without rebuilding the shot from scratch.

Why Choose This?

Conversational edit instructionsDescribe changes in natural language—swap an object, adjust wardrobe, or refresh music—without timeline-based manual cuts.
Explicit retain-and-change controlState what must stay and what must update so revisions stay localized to the intended region or element.
Temporal continuity preservedKeeps camera motion and scene flow coherent while applying the requested creative changes.
Native audio revisionsUpdate dialogue, music, or ambience with the picture so soundtrack changes land in the same MP4.
720p to 4K delivery tiersIterate edits at 720p or 1080p, then deliver 4K when the revised master needs higher clarity.
Landscape and portrait outputsChoose 16:9 or 9:16 so edited variants match the original distribution layout.

Parameters

Parameter	Requirement	Description
prompt	Required	String. Natural-language edit instruction covering changes and retained content; 1–20,000 characters after trimming.
video_urls	Required	Exactly one publicly accessible HTTP(S) video URL is required; missing, empty, or multiple entries are rejected. duration is not accepted; this endpoint imposes no source-video duration limit.
resolution	Optional	String. Output clarity tier; the Playground preselects 720p. Default`720p1080p4k`
aspect_ratio	Optional	String. Output framing; the Playground preselects 16:9. Default`16:99:16`

How to Use

Upload the source videoProvide one public MP4, MOV, or WebM URL as the footage you want to revise.
Describe changes and retained contentIdentify the edit target and desired change, then clearly list people, setting, or camera work to keep.
Add audio revision cues if neededSpecify dialogue, music mood, or ambience updates so soundtrack changes align with the picture.
Select resolutionPick 720p for iteration, 1080p for clearer delivery, or 4K for high-detail masters.
Choose aspect ratioSelect 16:9 or 9:16 to match the intended distribution layout.
Review the cost and runCheck the cost shown on the Run button, finish the source video and prompt, then click Run.
Preview and download the videoWhen the task finishes, preview the revised picture and audio, then select Download video to save the result.

Pricing

Billed per generation by output resolution, with native audio included. 1 credit = $0.005. Configure the source clip, edit prompt, resolution, and aspect ratio.

Usage	Rate	Details
720p / 1080p	300 credits	Default 720p costs 300 credits ($1.50).
4k	400 credits	4k costs 400 credits ($2.00).

Best Use Cases

Ad creative variantingSwap product colors, props, or on-screen talent details while keeping the approved camera path.
Dialogue and host retakesRefresh spoken lines or tone on an existing talking clip without reshooting the whole scene.
Soundtrack and ambience refreshReplace or swell music and effects to match a new brand mood on the same visual bed.
Localized scene adaptationsAdjust wardrobe, signage, or objects for regional versions while retaining core action.

Pro Tips

Identify the edit target by position, color, or clothing—for example the person in the blue jacket on the left.
Separate change and retain lines: Make the coat red. Keep the street setting and original camera movement.
Keep one primary edit goal per submission, then refine audio or appearance in a follow-up pass.
Describe music mood, entry point, and intensity so soundtrack revisions land on the intended beat.
Match aspect_ratio to the source layout when you need framing continuity across variants.

Usage notes

Gemini Omni Flash Video Edit requires prompt plus video_urls with exactly one public source video URL.
Configure resolution and aspect_ratio for delivery; credits follow the selected resolution tier.
Describe speech, music, or ambience changes in the prompt; native audio is included in the result.
After an API submission, save the returned task_id to query progress and retrieve the final media URL.

Related Models

Happy Horse 1.1 Text to VideoHappy Horse 1.1 generates 3–15 second videos at 720p or 1080p from text, a starting image, or reference images, with native synchronized audio and consistent subject identity.

Happy Horse 1.1 Image to VideoHappy Horse 1.1 generates 3–15 second videos at 720p or 1080p from text, a starting image, or reference images, with native synchronized audio and consistent subject identity.

Happy Horse 1.1 Reference to Video preview

Happy Horse 1.1 Reference to VideoHappy Horse 1.1 generates 3–15 second videos at 720p or 1080p from text, a starting image, or reference images, with native synchronized audio and consistent subject identity.

Gemini Omni 1.1 Flash Text-to-Video preview

Gemini Omni 1.1 Flash Text-to-VideoGoogle Gemini Omni 1.1 Flash video model with native audio, text-to-video, first/last-frame image-to-video, reference-image video, and 3–10 second video edit at 360p to 4K.

Gemini Omni 1.1 Flash Image-to-Video preview

Gemini Omni 1.1 Flash Image-to-VideoGoogle Gemini Omni 1.1 Flash video model with native audio, text-to-video, first/last-frame image-to-video, reference-image video, and 3–10 second video edit at 360p to 4K.

Gemini Omni 1.1 Flash Reference-to-Video preview

Gemini Omni 1.1 Flash Reference-to-VideoGoogle Gemini Omni 1.1 Flash video model with native audio, text-to-video, first/last-frame image-to-video, reference-image video, and 3–10 second video edit at 360p to 4K.

Gemini Omni 1.1 Flash Video Edit preview

Gemini Omni 1.1 Flash Video EditGoogle Gemini Omni 1.1 Flash video model with native audio, text-to-video, first/last-frame image-to-video, reference-image video, and 3–10 second video edit at 360p to 4K.

Wan 3.0 Text-to-VideoAlibaba Tongyi Lab's Wan 3.0 multimodal video model with separate text-to-video, first/last-frame image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds.

Wan 3.0 Image-to-VideoAlibaba Tongyi Lab's Wan 3.0 multimodal video model with separate text-to-video, first/last-frame image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds.

Wan 3.0 Reference-to-VideoAlibaba Tongyi Lab's Wan 3.0 multimodal video model with separate text-to-video, first/last-frame image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds.

Wan 3.0 Prime Text-to-VideoAlibaba Wan 3.0 Prime video model with text-to-video, image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds with optional synchronized audio.

Wan 3.0 Prime Image-to-VideoAlibaba Wan 3.0 Prime video model with text-to-video, image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds with optional synchronized audio.

Wan 3.0 Prime Reference-to-Video preview

Wan 3.0 Prime Reference-to-VideoAlibaba Wan 3.0 Prime video model with text-to-video, image-to-video, and reference-to-video workflows at 480p, 720p, or 1080p for 2 to 30 seconds with optional synchronized audio.

FLUX 3 Text to VideoBlack Forest Labs FLUX 3 video model with text-to-video, image-to-video, first-last-frame, extend-video, and keyframes workflows at 720p or 1080p for 5 to 20 seconds, with optional native audio.

FLUX 3 Image to VideoBlack Forest Labs FLUX 3 video model with text-to-video, image-to-video, first-last-frame, extend-video, and keyframes workflows at 720p or 1080p for 5 to 20 seconds, with optional native audio.

FLUX 3 First Last Frame to Video preview

FLUX 3 First Last Frame to VideoBlack Forest Labs FLUX 3 video model with text-to-video, image-to-video, first-last-frame, extend-video, and keyframes workflows at 720p or 1080p for 5 to 20 seconds, with optional native audio.

Gemini Omni Flash Video Edit API frequently asked questions

What is the Gemini Omni Flash Video Edit API?

Gemini Omni Flash Video Edit is a Google DeepMind multimodal model for conversational video revision. It edits one source clip through natural-language instructions, updating selected elements with native audio while preserving the footage and camera continuity you ask to keep, with output from 720p to 4K. Built on Gemini’s unified multimodal architecture, it localizes changes to the stated targets without rebuilding the entire shot from scratch. You can call it programmatically or try it from the playground above.

How do I specify retain regions in Gemini Omni Flash Video Edit?

Write separate retain instructions after the change request—for example Keep the street background and original camera move. Point to subjects by position, color, or clothing so the model knows what stays untouched.

How does Gemini Omni Flash Video Edit set clip length?

Output length follows the source video you submit. Provide video_urls and the edit prompt, then choose resolution and aspect ratio; credits follow the resolution tier in the Pricing section.

Can Gemini Omni Flash Video Edit refresh dialogue and music?

Yes. Describe new spoken lines, music mood, or ambience in the prompt so soundtrack revisions generate with the picture in the output MP4.

What source formats does Gemini Omni Flash Video Edit accept?

Submit exactly one public HTTP(S) video URL in video_urls using MP4, MOV, or WebM. The clip is the footage to revise; full field details are in the Parameters section.

When should Gemini Omni Flash Video Edit use 4K?

Choose 4K when the revised master needs sharper texture and lighting for final delivery. Iterate edits at 720p or 1080p first—same conversational workflow, lower credit cost.

How are Gemini Omni Flash Video Edit credits calculated?

Credits are charged per generation by resolution: 720p and 1080p cost 300 credits ($1.50), and 4K costs 400 credits ($2.00). See the Pricing section on this page.