Skip to main content

Interface: VisionResult

Defined in: packages/agentos/src/core/vision/types.ts:224

Aggregated result from the vision pipeline after all configured tiers have run. Contains the best extracted text, content classification, optional embeddings, and detailed per-tier breakdowns.

Example​

const result = await pipeline.process(imageBuffer);

// Best extracted text across all tiers
console.log(result.text);

// What kind of content was detected
console.log(result.category); // 'printed-text' | 'handwritten' | ...

// CLIP embedding for similarity search
if (result.embedding) {
  await vectorStore.upsert('images', [{ embedding: result.embedding }]);
}

// Inspect individual tier contributions
for (const tr of result.tierResults) {
  console.log(`${tr.tier} (${tr.provider}): ${tr.confidence}`);
}

Properties​

category​

category: VisionContentCategory

Defined in: packages/agentos/src/core/vision/types.ts:232

What kind of content was detected.

confidence​

confidence: number

Defined in: packages/agentos/src/core/vision/types.ts:229

Overall confidence score 0–1, taken from the winning tier.

durationMs​

durationMs: number

Defined in: packages/agentos/src/core/vision/types.ts:250

Total wall-clock processing time in milliseconds.

embedding?​

optional embedding: number[]

Defined in: packages/agentos/src/core/vision/types.ts:241

CLIP image embedding vector, when embedding tier is enabled.

layout?​

optional layout: DocumentLayout

Defined in: packages/agentos/src/core/vision/types.ts:244

Structured document layout, when Florence-2 ran.

regions?​

optional regions: VisionTextRegion[]

Defined in: packages/agentos/src/core/vision/types.ts:247

Bounding boxes for detected text regions from the winning tier.

text​

text: string

Defined in: packages/agentos/src/core/vision/types.ts:226

Best extracted text (from OCR, handwriting, or vision description).

tierResults​

tierResults: VisionTierResult[]

Defined in: packages/agentos/src/core/vision/types.ts:238

Detailed results from each tier that ran, ordered by execution.

tiers​

tiers: VisionTier[]

Defined in: packages/agentos/src/core/vision/types.ts:235

Which tier(s) contributed to the final result.