Google Gemini Omni 1.2 Flash Image To Video API Documentation

Google Gemini Omni 1.2 Flash Image To Video API Documentation

Playground

Gemini Omni 1.2 Flash Image-to-Video animates a start image into short AI videos, with optional end-frame guidance and synchronized audio for social content, creative storytelling, marketing videos, and production workflows. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Gemini Omni 1.2 Flash Image-to-Video animates a start image into a video with synchronized audio. You can optionally add a final frame to guide the ending of the shot, making it useful for controlled transitions, reveal sequences, camera movement, and other prompt-guided image-to-video workflows.

Why Choose This?

• Start-frame image animation
  Turn a single image into a video with motion, scene progression, and synchronized audio.

• Optional end-frame guidance
  Add last_image when you want more control over how the shot ends.

• Smooth transition generation
  Useful for camera moves, reveal shots, visual transformations, and controlled motion between two keyframes.

• Draft-to-production workflow
  Use 360p for quick testing, then move to 1080p or 4k for higher-quality output.

• Synchronized audio
  Generate motion and matching audio together in one workflow.

• Flexible output formats
  Supports both 16:9 and 9:16 for landscape and vertical video creation.

Parameters

How to Use

1. Upload the start image — Provide the image that should define the first frame.
2. Add an end image optional — Use last_image when you want to guide the final frame or overall transition direction.
3. Write your prompt — Describe the motion, camera movement, transition behavior, scene changes, style, and audio.
4. Choose aspect ratio — Use 16:9 for landscape video or 9:16 for vertical content.
5. Select resolution — Use 360p for drafts, 720p for standard output, or 1080p / 4k for higher-quality results.
6. Set duration — Choose a duration from 3 to 15 seconds.
7. Submit — Generate the final video with synchronized audio.

Pricing

Pricing is based on generated video duration and selected resolution.

Example Costs

Best Use Cases

• Image animation — Turn still images into short motion clips with audio.
• Controlled transitions — Use image and last_image to guide the beginning and end of a shot.
• Reveal and transformation scenes — Create visual change between two defined states.
• Social media content — Generate vertical or landscape motion content from static artwork or product images.
• Creative prototyping — Quickly test motion, pacing, and shot progression from a single keyframe.
• Marketing and campaign assets — Animate posters, product visuals, and branded images into short videos.

Pro Tips

• Use a clear, high-quality start image for better subject consistency.
• Add last_image when the final pose, composition, or scene state matters.
• Describe both what moves and how it moves in the prompt.
• Include camera instructions such as push-in, orbit, pan, or pull-back for more controlled results.
• Mention ambience, music, dialogue, or sound effects directly when audio guidance matters.
• Start with 360p for fast iteration, then move to 1080p or 4k for final output.
• Use shorter durations for quick tests and longer durations when the transition needs more time to develop.

Notes

• When last_image is provided, choose a duration of 3–10 seconds. Without last_image, durations of 3–15 seconds are supported.

Related Models

• Gemini Omni 1.2 Flash Text-to-Video — Generate videos directly from text prompts with synchronized audio.
• Gemini Omni 1.2 Flash Reference-to-Video — Generate videos using reference media for stronger visual guidance.
• Gemini Omni 1.2 Flash Video Edit — Edit existing videos with prompt instructions.

Authentication

For authentication details, please refer to the Authentication Guide.

API Endpoints

Submit Task & Query Result

set -euo pipefail

export WAVESPEED_API_KEY="your-api-key"

REQUEST_BODY=$(cat <<'JSON'
{
  "image": "https://interactive-examples.mdn.mozilla.net/media/cc0-images/painted-hand-298-332.jpg",
  "prompt": "A cinematic ocean wave at sunrise, highly detailed",
  "aspect_ratio": "16:9",
  "resolution": "720p",
  "duration": 8
}
JSON
)

# 1. Submit the prediction once. Never retry this request automatically:
# a retry after an unclear failure can create a second, billed task.
CHECK='The task may already exist. Check History or the predictions list before submitting again.'
if ! SUBMIT_RESPONSE=$(curl --silent --show-error --fail-with-body \
  --connect-timeout 10 --max-time 60 \
  -X POST "https://api.wavespeed.ai/api/v3/google/gemini-omni-1.2-flash/image-to-video" \
  -H "Authorization: Bearer ${WAVESPEED_API_KEY}" \
  -H "Content-Type: application/json" \
  -d "${REQUEST_BODY}"); then
  printf 'Submit failed. %s\n' "${CHECK}" >&2
  exit 1
fi

TASK=$(printf '%s' "${SUBMIT_RESPONSE}" | jq 'if type == "object" and has("data") then .data else . end')
PREDICTION_ID=$(printf '%s' "${TASK}" | jq -r '.id // empty')
if [ -z "${PREDICTION_ID}" ]; then
  printf 'Submission response did not contain a prediction id. %s\n' "${CHECK}" >&2
  exit 1
fi
RESULT_URL="https://api.wavespeed.ai/api/v3/predictions/${PREDICTION_ID}/result"

# 2. Poll until the prediction reaches a final status.
BODY_FILE=$(mktemp)
trap 'rm -f "${BODY_FILE}"' EXIT
while true; do
  # Transient errors (429, 5xx) are retried, honoring Retry-After.
  HTTP_CODE=$(curl --silent --show-error --fail-with-body \
    --connect-timeout 10 --max-time 60 --retry 5 --retry-delay 2 \
    --output "${BODY_FILE}" --write-out '%{http_code}' \
    "${RESULT_URL}" \
    -H "Authorization: Bearer ${WAVESPEED_API_KEY}" || true)
  case "${HTTP_CODE}" in
    2??) ;;
    000|429|5??)  # dropped connection or temporary error: keep polling
      printf 'Task %s: HTTP %s, polling again in 10s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      sleep 10; continue ;;
    *) printf 'Task %s: HTTP %s\n' "${PREDICTION_ID}" "${HTTP_CODE}" >&2
      cat "${BODY_FILE}" >&2; exit 1 ;;
  esac
  RESULT=$(jq 'if type == "object" and has("data") then .data else . end' "${BODY_FILE}")
  STATUS=$(printf '%s' "${RESULT}" | jq -r '.status // empty')

  case "${STATUS}" in
    completed) printf '%s\n' "${RESULT}" | jq '.outputs'; break ;;
    failed|cancelled|timeout|deleted)
      printf 'Task %s ended with status %s\n' "${PREDICTION_ID}" "${STATUS}" >&2
      printf '%s\n' "${RESULT}" | jq . >&2; exit 1 ;;
    *) sleep 2 ;;
  esac
done

Parameters

Task Submission Parameters

Request Parameters

Every model also accepts the global request options enable_sync_mode and enable_base64_output; see their pages for details.

Response Parameters

Result Request Parameters

Result Response Parameters