/**
 * Local Sentence-BERT Embeddings
 *
 * Runs embeddings locally using a Python service
 * Faster and free (after setup), but requires Python
 */
import type { Embedder } from '../embeddingPipeline.js';
export interface LocalEmbedderConfig {
    serviceUrl?: string;
    timeout?: number;
}
/**
 * Local embedder that calls a Python/FastAPI service
 *
 * Setup instructions:
 * 1. Install: pip install sentence-transformers fastapi uvicorn
 * 2. Run server: python embedding-server.py
 * 3. Use this embedder
 */
export declare class LocalEmbedder implements Embedder {
    private serviceUrl;
    private timeout;
    constructor(config?: LocalEmbedderConfig);
    embedText(text: string): Promise<Float32Array>;
    embedDiff(diff: string): Promise<Float32Array>;
    private getEmbedding;
}
/**
 * Python server code (save as embedding-server.py):
 *
 * ```python
 * from fastapi import FastAPI
 * from sentence_transformers import SentenceTransformer
 * from pydantic import BaseModel
 * import uvicorn
 *
 * app = FastAPI()
 * model = SentenceTransformer('all-MiniLM-L6-v2')  # 384 dimensions, fast
 *
 * class EmbedRequest(BaseModel):
 *     text: str
 *     type: str = "text"
 *
 * @app.post("/embed")
 * async def embed(request: EmbedRequest):
 *     embedding = model.encode(request.text)
 *     return {"embedding": embedding.tolist()}
 *
 * if __name__ == "__main__":
 *     uvicorn.run(app, host="0.0.0.0", port=8000)
 * ```
 *
 * Run with: python embedding-server.py
 */
//# sourceMappingURL=local.d.ts.map