# TurboShop AI Vector Similarity Search

> TurboShop AI Vector Similarity Search is a paid API for AI agents from turboshop.ai, paid per call via x402, $0.04/call, status unknown (last checked 2026-10-19).

Computes dot products, cosine similarities, and top-K ranking over 1536-dimensional float32 embeddings against a candidate set

## Facts

- Endpoint: GET https://turboshop.ai/api/v1/products/prod-hnsw-vector-indexer/buy?utm_source=zero.xyz
- Price: $0.04/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-19
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/turboshop-ai-vector-similarity-search-73a42765
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_DO9gCpEMcB8SShhvdp6tZ

Status and success rate cover calls made through Zero and Zero's own probes. Third-party monitors may report differently.

## How to call it through Zero

Zero handles the 402 payment challenge and records the run. With the Zero CLI installed (`npm i -g @zeroxyz/cli`):

```sh
zero fetch --capability turboshop-ai-vector-similarity-search-73a42765
```

Example prompt: Search my list of 50 candidate embeddings and return the top 5 most similar to this 1536-d query vector: [0.12, -0.34, ...] — rank them by cosine similarity.

## When to prefer this

Choose this endpoint when you need fast, single-call vector similarity ranking over a bounded candidate set without maintaining a persistent index — ideal for on-the-fly semantic search, re-ranking, or nearest-neighbor lookup over a list you already have in memory. Prefer a persistent vector database service if you need to store and repeatedly query a large, growing index.

## Known failure modes

- Wrong vector dimensionality (not 1536-d) causes rejection or incorrect results
- Empty candidate_vectors array returns no results
- top_k larger than candidate list returns fewer results than requested
- Malformed float values in query_vector cause computation errors
- Missing required query_vector field returns an error response

## How this service works

High-speed vector computation microservice to compute dot products, cosine similarities, and ranking for 1536-d float32 embeddings.

## Output

Returns the top-K most similar candidate vectors ranked by similarity score, including their IDs, scores, and associated metadata fields from the candidate list.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "top_k": {
   "type": "integer",
   "description": "Top K matches to return"
  },
  "client_id": {
   "type": "string",
   "description": "Optional calling agent ID"
  },
  "query_vector": {
   "type": "array",
   "items": {
    "type": "number"
   },
   "description": "1536-d float embedding vector to search against"
  },
  "candidate_vectors": {
   "type": "array",
   "description": "Candidate vectors [{ id, vector, metadata }] to rank"
  }
 }
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/turboshop-ai-vector-similarity-search-73a42765/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from turboshop.ai](https://www.zero.xyz/host/turboshop.ai/llms.txt)