Interface VisionTierResult

Result from a single processing tier.

Each tier that runs produces one of these, regardless of whether the pipeline ultimately uses its output or prefers a higher-confidence alternative from another tier.

interface VisionTierResult {
    tier: VisionTier;
    provider: string;
    text: string;
    confidence: number;
    durationMs: number;
    regions?: VisionTextRegion[];
}

Properties

Which tier produced this result.

provider: string

Provider name within the tier (e.g. 'paddle', 'tesseract', 'openai').

text: string

Extracted or generated text content.

confidence: number

Confidence score from 0 (no confidence) to 1 (certain).

durationMs: number

Wall-clock processing time in milliseconds.

regions?: VisionTextRegion[]

Bounding boxes for detected text regions, when available.