Bytedance Video Upscaler

POST/v1/run

Bytedance Video Upscaler upscales videos to 4K resolution with AI-powered enhancement. Supports multiple quality tiers (fast, standard, pro), frame rate interpolation to 60fps, and presets optimized for UGC, short series, AIGC, and old film restoration.

Request body

Submit an async generation request. The model field selects the model; other fields are model-specific input parameters.

stringmodelrequired

Model identifier. Set to bytedance/upscaler/video.

Default: bytedance/upscaler/video

Optional<string>video

The URL of the video to upscale.

Optional<string>enhancement_preset

The enhancement preset optimized for specific video scenarios. 'general' is a general-purpose template, 'ugc' targets user-generated short videos, 'short_series' is for short dramas, 'aigc' is for AI-generated content, and 'old_film' is for classic film restoration.

Allowed values: general, ugc, short_series, aigc, old_film

Default: general

Optional<number>scale_ratio

The scaling ratio for the output video resolution. When set, overrides target_resolution and scales the input resolution by this factor (e.g., 2.0 doubles the resolution). Range: 1.1 to 10.0. Please note that this is valid only up to 4k resolution, and trying to scale beyond 4k will result in an error. (4k is defined as having atotal pixel count of 3840x2160).

Range: 1.1 to 10

Optional<string>target_resolution

The target resolution of the video to upscale.

Allowed values: 1080p, 2k, 4k

Default: 1080p

Optional<string>target_fps

The target FPS of the video to upscale.

Allowed values: 30fps, 60fps

Default: 30fps

Optional<string>enhancement_tier

The enhancement quality tier. 'fast' provides essential upscaling with good speed, 'standard' uses adaptive algorithms for better visual texture, 'pro' uses large-model restoration for cinematic quality (longer processing time), and 10 times the cost of `standard` and `fast`.

Allowed values: fast, standard, pro

Default: standard

Optional<string>fidelity

The enhancement intensity. 'high' applies mild enhancement while keeping visual texture close to the source video. 'medium' provides a balanced image quality enhancement.

Allowed values: high, medium

Default: high

Response Schema

The submit endpoint returns an accepted generation task. Poll the result endpoint with the returned id for terminal outputs or errors.

Optional<string>error

Error message if the task failed. Empty on success.

stringidrequired

Unique identifier for the generation task.

Optional<string>model

Model ID used for the prediction.

Optional<array>outputs

Array of generated content. Empty when status is not completed.

stringstatusrequired

Status of the task: pending, running, completed, failed, or timeout.

Allowed values: pending, running, completed, failed, timeout

Model capabilities

array<string>capability_tagsrequired

Capabilities declared by the model registry.

Default: video-to-video

stringexecution_moderequired

Execution mode declared by the model registry.

Default: async

# 1. Submit a generation request
curl -X POST https://api.sandbase.ai/v1/run \
  -H "Authorization: Bearer $SANDBASE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
  "model": "bytedance/upscaler/video",
  "video": "https://static.sandbase.ai/examples/bytedance/upscaler/video/input_video_0.mp4"
}'

# 2. If the response is still running, poll the returned id
curl https://api.sandbase.ai/v1/run/run_abc123 \
  -H "Authorization: Bearer $SANDBASE_API_KEY"

Response202 Accepted

{
  "id": "run_abc123",
  "model": "bytedance/upscaler/video",
  "status": "running",
  "outputs": []
}