batch_process_embeddings
Process large content batches into embeddings through automated workflow: ingest content, convert to JSONL, upload, create batch job, poll completion, and download 1536-dimensional vector results with metadata.
Instructions
COMPLETE EMBEDDINGS WORKFLOW - End-to-end embeddings batch processing. WORKFLOW: 1) Ingests content, 2) Queries user for task type (or auto-recommends), 3) Converts to JSONL, 4) Uploads, 5) Creates batch job, 6) Polls until complete, 7) Downloads results. BEST FOR: Simple one-call embeddings generation. RETURNS: Embeddings array (1536-dimensional vectors) with metadata.
Input Schema
| Name | Required | Description | Default |
|---|---|---|---|
| inputFile | Yes | Path to content file | |
| taskType | No | Embedding task type (omit to get interactive prompt) | |
| model | No | Embedding model | gemini-embedding-001 |
| outputLocation | No | Output directory for results | |
| pollIntervalSeconds | No | Seconds between status checks |
Input Schema (JSON Schema)
{
"properties": {
"inputFile": {
"description": "Path to content file",
"type": "string"
},
"model": {
"default": "gemini-embedding-001",
"description": "Embedding model",
"enum": [
"gemini-embedding-001"
],
"type": "string"
},
"outputLocation": {
"description": "Output directory for results",
"type": "string"
},
"pollIntervalSeconds": {
"default": 30,
"description": "Seconds between status checks",
"minimum": 10,
"type": "number"
},
"taskType": {
"description": "Embedding task type (omit to get interactive prompt)",
"enum": [
"SEMANTIC_SIMILARITY",
"CLASSIFICATION",
"CLUSTERING",
"RETRIEVAL_DOCUMENT",
"RETRIEVAL_QUERY",
"CODE_RETRIEVAL_QUERY",
"QUESTION_ANSWERING",
"FACT_VERIFICATION"
],
"type": "string"
}
},
"required": [
"inputFile"
],
"type": "object"
}