Developers

Documentation

Embedder provides a unified API for embedding generation, vector indexing, and semantic retrieval. This guide covers authentication, core endpoints, and SDK installation.

Authentication

All requests require a Bearer token. Generate API keys from the console after your workspace is provisioned.

Authorization: Bearer $EMBEDDER_KEY
Content-Type: application/json

Quick start

# Install SDK
pip install embedder

# Create an embedding
curl -X POST https://api.embedder.lat/v1/embed \
  -H "Authorization: Bearer $EMBEDDER_KEY" \
  -d '{"input": "your text here", "model": "embed-v2"}'

# Query a vector index
curl -X POST https://api.embedder.lat/v1/index/production/query \
  -H "Authorization: Bearer $EMBEDDER_KEY" \
  -d '{"vector": [...], "top_k": 10, "filter": {"region": "latam"}}'

Core endpoints

Method	Endpoint	Description
POST	/v1/embed	Generate embeddings
POST	/v1/index/{id}/upsert	Insert or update vectors
POST	/v1/index/{id}/query	Similarity search
DELETE	/v1/index/{id}/vectors	Remove vectors by ID
GET	/v1/index/{id}/stats	Index statistics

SDKs

Python — pip install embedder
TypeScript — npm install @embedder/sdk
Go — go get github.com/embedder-lat/sdk-go

Need access? Request early access or email developers@embedder.lat.