Vector Stores

GET    /v1/vector_stores
POST   /v1/vector_stores
GET    /v1/vector_stores/:vector_store_id
POST   /v1/vector_stores/:vector_store_id
DELETE /v1/vector_stores/:vector_store_id

GET    /v1/vector_stores/:vector_store_id/files
POST   /v1/vector_stores/:vector_store_id/files
GET    /v1/vector_stores/:vector_store_id/files/:file_id
DELETE /v1/vector_stores/:vector_store_id/files/:file_id

Requires Authorization: Bearer sk-bve-YOUR_KEY.

Vector stores hold chunked, embedded representations of files for use with the Assistants file search tool.

Create a vector store

curl -X POST https://api.bve.me/v1/vector_stores \
  -H "Authorization: Bearer sk-bve-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -H "OpenAI-Beta: assistants=v2" \
  -d '{
    "name": "My Knowledge Base"
  }'

Response:

{
  "id": "vs_abc123",
  "object": "vector_store",
  "created_at": 1716288000,
  "name": "My Knowledge Base",
  "status": "completed",
  "file_counts": {
    "in_progress": 0,
    "completed": 0,
    "failed": 0,
    "cancelled": 0,
    "total": 0
  }
}

Add files to a vector store

# First, upload the file
curl https://api.bve.me/v1/files \
  -H "Authorization: Bearer sk-bve-YOUR_KEY" \
  -F purpose="assistants" \
  -F file="@document.pdf"

# Then attach to the vector store
curl -X POST https://api.bve.me/v1/vector_stores/vs_abc123/files \
  -H "Authorization: Bearer sk-bve-YOUR_KEY" \
  -H "Content-Type: application/json" \
  -H "OpenAI-Beta: assistants=v2" \
  -d '{ "file_id": "file-abc123" }'

OpenAI SDK

import OpenAI from 'openai';
import fs from 'fs';

const client = new OpenAI({
  apiKey: 'sk-bve-YOUR_KEY',
  baseURL: 'https://api.bve.me/v1',
});

// Create vector store
const vectorStore = await client.beta.vectorStores.create({
  name: 'My Knowledge Base',
});

// Upload and attach a file
const file = await client.files.create({
  file: fs.createReadStream('document.pdf'),
  purpose: 'assistants',
});

await client.beta.vectorStores.files.create(vectorStore.id, {
  file_id: file.id,
});

// Use with an assistant
const assistant = await client.beta.assistants.create({
  name: 'Search Assistant',
  model: 'gpt-4o',
  tools: [{ type: 'file_search' }],
  tool_resources: {
    file_search: {
      vector_store_ids: [vectorStore.id],
    },
  },
});

Notes

All vector store operations proxy directly to Fuelix.
Include OpenAI-Beta: assistants=v2 on requests.
Files must first be uploaded via Files API before attaching to a vector store.