Gemini Omni 1.1 Flash Video Edit API

google/gemini-omni-1.1-flash/video-edit

Gemini Omni 1.1 Flash (Video Edit) edits 3–10 second source videos through natural-language instructions, with reference-image guidance, native audio, and 360p to 4K output. Specify what to change and what to retain to develop new versions of existing footage through character, object, visual style, or sound edits.

Input

Prompt

568/20000

Source video

Drag and drop files, paste from clipboard, or provide a URL.

Reference images

Drag and drop files, paste from clipboard, or provide a URL.

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 1.1 Flash Text-to-Video Gemini Omni 1.1 Flash Image-to-Video Gemini Omni 1.1 Flash Reference-to-Video

Examples

Replace only the woman's burgundy jacket with the mustard-yellow raincoat shown in the reference image. Match its mustard fabric, brown toggle fastenings, lower patch pockets and folded hood; the hood stays down. Make the coat fit her naturally throughout the moving shot. Preserve her exact face, dark curly hair, hands, dark trousers, walking action and timing. Preserve the coastal path, cliffs, ocean, overcast light and original camera movement. Keep the existing wind, surf and footsteps without adding speech or music. No new people, text, logos or advertising.

Remove the cyclist and the entire bicycle from every frame. Reconstruct the stone paving and any occluded background naturally. Preserve the blue-and-white tiled fountain, flowing water, pigeons, whitewashed buildings, sunlight, original locked camera and timing. Do not replace the cyclist with another person or object. Keep fountain water and pigeon ambience, remove any bicycle sound, add no music or speech. No text, logos or advertising.

Transform the entire shot into a hand-painted oil-on-canvas animation: visible expressive impasto brushstrokes, warm ochre and burnt-umber pigments, soft painterly edges and tactile canvas texture. Preserve the original composition, two hands and their exact finger actions, turning wheel, clay bowl shape, workshop shelves, left-side light and timing. The brush texture must remain coherent across frames without flickering. Retain the original quiet wheel motor and wet-clay friction, add no music or speech. No new objects, text, logos or advertising.

REST API

Quick Start

Authenticate with the API, submit the inputs and instructions, then retrieve the video 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 inputs and settings for this endpoint using the request example, then save the returned task_id to query generation progress and results.

Submit request

REQUEST_BODY=$(cat <<'JSON'
{
  "model": "google/gemini-omni-1.1-flash/video-edit",
  "input": {
    "prompt": "Replace only the woman's burgundy jacket with the mustard-yellow raincoat shown in the reference image. Match its mustard fabric, brown toggle fastenings, lower patch pockets and folded hood; the hood stays down. Make the coat fit her naturally throughout the moving shot. Preserve her exact face, dark curly hair, hands, dark trousers, walking action and timing. Preserve the coastal path, cliffs, ocean, overcast light and original camera movement. Keep the existing wind, surf and footsteps without adding speech or music. No new people, text, logos or advertising.",
    "reference_video_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-1.1-flash/video-edit/v2/01/input-source.mp4"
    ],
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "reference_image_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-1.1-flash/video-edit/v2/01/input-reference.webp"
    ]
  }
}
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-1.1-flash/video-edit",
  "input": {
    "prompt": "Replace only the woman's burgundy jacket with the mustard-yellow raincoat shown in the reference image. Match its mustard fabric, brown toggle fastenings, lower patch pockets and folded hood; the hood stays down. Make the coat fit her naturally throughout the moving shot. Preserve her exact face, dark curly hair, hands, dark trousers, walking action and timing. Preserve the coastal path, cliffs, ocean, overcast light and original camera movement. Keep the existing wind, surf and footsteps without adding speech or music. No new people, text, logos or advertising.",
    "reference_video_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-1.1-flash/video-edit/v2/01/input-source.mp4"
    ],
    "resolution": "720p",
    "aspect_ratio": "16:9",
    "reference_image_urls": [
      "https://cdn.vidgo.ai/apis/models/google/gemini-omni-1.1-flash/video-edit/v2/01/input-reference.webp"
    ]
  }
}
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. Prepare the inputs for this task and configure the output.

Field	Type	Required	Default	Description
prompt	string	Yes	—	1–20,000 characters after trimming.
reference_video_urls	string[]	Yes	—	Exactly one public video URL. Source duration must be 3–10 seconds.
reference_image_urls	string[]	No	—	Omit this field or provide an empty array; accepts up to 5 public image URLs for appearance or style guidance.
resolution	string	No	720p	360p, 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	Generation progress from 0 to 100, when available.
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 inputs, parameter values, and available credits, then make the indicated adjustments before submitting again.
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	Combine a source clip with an edit instruction, with up to 5 optional images to guide the target appearance.
Source video	MP4 / MOV / WebM	Exactly one clip, 3–10 seconds, up to 100 MiB.
Output	Video with native audio	The endpoint returns an asynchronous task ID.
Resolution	360p / 720p / 1080p / 4k	Default is 720p.
Aspect ratio	16:9 / 9:16	Default is 16:9.
Billing basis	Per generation	360p, 720p, and 1080p cost 120 credits. 4k costs 180 credits.

View Gemini Omni 1.1 Flash Video Edit API documentation

Gemini Omni 1.1 Flash Video Edit

Gemini Omni 1.1 Flash Video Edit applies natural-language edits to a 3–10 second source video, with up to 5 optional reference images to guide the target appearance. Describe changes to people, objects, visual details, or sound and identify the content to retain to create a new version of existing footage.

Why Choose This?

Natural-language editingDescribe the target and desired change, such as a clothing adjustment or musical addition, to turn creative revisions into clear editing instructions.
Character and object editsDescribe additions, removals, or replacements involving a specific person or object to develop new versions of an existing scene.
Reference-guided appearanceAdd up to 5 reference images to show target clothing, objects, or visual style and give the edit a concrete visual direction.
Defined retained contentDistinguish the edit target from content to retain, organizing revisions around the source scene, camera work, and action.
Music and sound revisionsDescribe musical mood, sound effects, and audio changes to give existing footage a new atmosphere in a video with native audio.
Output choices for deliveryChoose 360p, 720p, 1080p, or 4K with landscape or portrait framing to configure the edited result for further editing and presentation.

Parameters

Parameter	Requirement	Description
prompt	Required	String. Natural-language edit instruction; 1–20,000 characters after trimming.
reference_video_urls	Required	Array containing one public video URL as the footage to edit. Use a 3–10 second MP4, MOV, or WebM clip, up to 100 MiB.
reference_image_urls	Optional	Omit this field or provide an empty array; accepts up to 5 public image URLs for appearance or style guidance.
resolution	Optional	String. Sets output resolution; the Playground preselects 720p. Default`720p360p1080p4k`
aspect_ratio	Optional	String. Controls output framing; the Playground preselects 16:9. Default`16:99:16`

How to Use

Upload a source videoAdd one 3–10 second MP4, MOV, or WebM clip, up to 100 MiB, as the footage to edit.
Describe changes and retained contentIdentify the edit target and desired change, then describe the people, setting, or camera work you want to retain.
Add optional reference imagesTo show target clothing, objects, or styling, add up to 5 JPEG, PNG, or WebP images, up to 30 MiB each.
Choose a resolutionChoose 360p, 720p, 1080p, or 4K, with 720p selected by default, to match the output specifications for editing or presentation.
Choose an aspect ratioChoose 16:9 landscape or 9:16 portrait, with 16:9 selected by default, and frame the subject for the intended layout.
Review the cost and runReview the cost shown on the Run button, complete the required uploads and prompt, then click Run.
Preview and download the videoWhen the task finishes, preview the video and sound in the output panel, then select Download video to save the result.

Pricing

Billed per generation based on output resolution, with native audio included in the result. 1 credit = $0.005.

Usage	Rate	Details
360p / 720p / 1080p	120 credits / generation	Default 720p costs 120 credits ($0.60).
4k	180 credits / generation	4k costs 180 credits ($0.90).

Best Use Cases

Scene element revisionsIdentify a person or object in existing footage and describe the change to create a revised story shot or product presentation.
Product appearance versionsCombine source footage with product reference images to describe a target appearance for a new visual direction.
Music and sound revisionsDescribe musical style, emotional progression, and effect timing to create a new soundtrack direction for an opening sequence or brand presentation.
Brand asset adaptationsProvide new clothing, scene styling, or object references to adapt an existing short clip for different brand themes.

Pro Tips

Identify the edit target by position, color, or clothing, such as “the person in the blue sweater on the left,” to point to a specific subject.
Describe changes and retained content separately, such as “Make the coat red. Keep the street setting and original camera movement.”
Use reference images to show target clothing, objects, or visual style and explain which edit each image guides.
Specify the mood, rhythm, and entry point of music, such as strings gradually swelling as a character turns.
Organize each submission around one clear editing goal, then review the result before refining the relevant action, appearance, or sound direction.

Usage notes

Gemini Omni 1.1 Flash Video Edit uses one 3–10 second source video and an edit instruction through prompt and reference_video_urls. Add up to 5 optional reference images to guide the target appearance and choose the output resolution and aspect ratio.
Describe speech, music, or ambient sound in the prompt; native audio is included in the result.
Use publicly accessible HTTP(S) URLs for API media inputs so the service can retrieve the files.
Save the task_id returned by an API submission to query progress and retrieve the result.

Related Models

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.

MiniMax H3 Max Turbo Text to Video preview

MiniMax H3 Max Turbo Text to VideoMiniMax H3 Max Turbo Text to Video turns text into coherent 5–15 second videos with synchronized native stereo audio and fast generation. It follows physical motion while rendering subtle changes in light and shadow.

MiniMax H3 Max Turbo Image to Video preview

MiniMax H3 Max Turbo Image to VideoMiniMax H3 Max Turbo Text to Video turns text into coherent 5–15 second videos with synchronized native stereo audio and fast generation. It follows physical motion while rendering subtle changes in light and shadow.

MiniMax H3 Max Text to VideoMiniMax H3 Max Text to Video turns text prompts into cinematic video with native stereo audio, faster-than-real-time generation, and precise instruction following. It renders complex camera work and physical motion with fine detail across multiple resolutions and common aspect ratios.

MiniMax H3 Max Image to VideoMiniMax H3 Max Text to Video turns text prompts into cinematic video with native stereo audio, faster-than-real-time generation, and precise instruction following. It renders complex camera work and physical motion with fine detail across multiple resolutions and common aspect ratios.

MiniMax H3 Max Reference to Video preview

MiniMax H3 Max Reference to VideoMiniMax H3 Max Text to Video turns text prompts into cinematic video with native stereo audio, faster-than-real-time generation, and precise instruction following. It renders complex camera work and physical motion with fine detail across multiple resolutions and common aspect ratios.

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.

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.

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.

Gemini Omni 1.1 Flash Video Edit API frequently asked questions

What is the Gemini Omni 1.1 Flash Video Edit API?

Gemini Omni 1.1 Flash Video Edit is a Google model for instruction-based video editing. It applies localized visual and acoustic modifications to 3–10 second source clips using natural language, supporting up to 5 reference images for precise aesthetic guidance and outputting native audio. Built on Gemini's conversational multimodal architecture, it uses the source characters, camera trajectories, and temporal rhythm as context while editing the target elements. You can call it programmatically or try it from the playground above.

What video elements can Gemini Omni 1.1 Flash Video Edit modify?

You can modify character apparel textures and colors, add or remove specific scene objects, alter backgrounds and weather conditions, or apply global artistic style transfers. Natural-language instructions direct the edits without requiring manual rotoscoping or frame-by-frame masking.

Can Gemini Omni 1.1 Flash Video Edit replace background music independently?

Yes. The model can revise the audio track in tandem with visual edits or independently. For instance, prompting to retain ambient room tone while swapping background music for a suspenseful orchestral score produces a newly scored soundtrack aligned with on-screen action.

Does Gemini Omni 1.1 Flash Video Edit support uploading reference images?

Yes. You can attach up to 5 reference images to illustrate the target style or item appearance. For example, instructing the model to replace a backpack with the leather bag shown in Reference Image 1 anchors the modification to concrete visual details.

Does Gemini Omni 1.1 Flash Video Edit preserve original camera movements?

Yes. When performing localized modifications, the model uses the source camera path, panning speed, and timing as context when editing the specified content, supporting continuity with the original shot.

What are the source video duration limits for Gemini Omni 1.1 Flash Video Edit?

Input source videos should be between 3 and 10 seconds in duration and formatted as MP4, MOV, or WebM. Keeping clips within this window provides the optimal balance of temporal consistency analysis and processing responsiveness.

How can I prevent unwanted changes in Gemini Omni 1.1 Flash Video Edit?

Use a dual-instruction prompt structure that specifies both modifications and retentions. For example, write "Change the man's coat to a dark grey trench coat. Keep original face, hair, camera movement, and background pedestrians unchanged." Explicit boundary instructions minimize collateral modifications.