Bytedance Seedance V1 Lite I2V 480p

Bytedance Seedance V1 Lite I2V 480p

Playground

ByteDance Seedance V1 Lite is a 480p Image-to-Video model for coherent multi-shot videos with smooth motion and precise prompts. Ready-to-use REST inference API, best performance, no coldstarts, affordable pricing.

Features

Seedance v1 Lite I2V 480p — bytedance/seedance-v1-lite-i2v-480p

Seedance v1 Lite I2V 480p generates short videos from a single reference image plus a motion-focused prompt. Upload a starting image, describe the action and camera beats, and the model animates the scene while keeping the input image as the visual anchor. You can also provide an optional last_image to guide how the clip ends, and enable camera_fixed when you want motion without camera movement.

Key capabilities

• Image-to-video generation anchored to a reference image (480p output)
• Lightweight, fast iteration for short clips and quick variations
• Optional last_image support for smoother ending continuity
• Optional camera_fixed mode for stable framing
• Seed control for reproducible takes

Use cases

• Action shots from a single keyframe (sports, skating, dance, gestures)
• Social prototypes and quick story beats at low cost
• Start-to-end guided shots using last_image for more predictable endings
• Product/object animation with stable framing (camera_fixed)
• Generating multiple variations quickly by changing seed

Pricing

Inputs

• image (required): starting reference image (first-frame anchor)
• prompt (optional/required by UI): motion and scene direction
• last_image (optional): end-frame reference to guide the finishing state

Parameters

• prompt: describes subject action, environment motion, and camera direction
• image: input image (upload or URL)
• last_image: optional end reference image (upload or URL)
• duration: video length in seconds
• aspect_ratio: output aspect ratio selection
• camera_fixed: whether to lock the camera position
• seed: random seed (-1 for random; fixed value for reproducible results)
• safety checker: enabled filtering for safer generations (if available in your UI)

Prompting guide (I2V)

Write prompts like a director’s beat sheet:

• Subject action: what changes over time (glide, jump, turn, land, celebrate)
• Camera: follow, dolly, pan, slow push-in (or keep fixed)
• Environment: motion blur cues, dust, wind, crowd movement, lighting shifts
• Continuity: explicitly say what must remain consistent (subject identity, outfit, setting)

Example prompts

• A skateboarder performs a smooth boardslide along the ledge in a sunny skatepark. Camera follows from a low wide angle, then gently pushes in as he lands. Natural motion blur, crisp daylight, energetic vibe.
• A runner sprints past the camera, dust kicks up, shallow depth of field, camera pans to follow, dynamic sports cinematography, 5 seconds.
• A dancer spins and stops on beat, stage lights flicker softly, camera slow orbit, clean motion, 5 seconds.

# Submit the task
curl --location --request POST "https://api.wavespeed.ai/api/v3/bytedance/seedance-v1-lite-i2v-480p" \
--header "Content-Type: application/json" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}" \
--data-raw '{
    "duration": 5,
    "camera_fixed": false,
    "seed": -1
}'

# Get the result
curl --location --request GET "https://api.wavespeed.ai/api/v3/predictions/${requestId}/result" \
--header "Authorization: Bearer ${WAVESPEED_API_KEY}"