# ZeroReader BGE-M3 Multilingual Embeddings

> ZeroReader BGE-M3 Multilingual Embeddings is a paid API for AI agents from api.zeroreader.com, paid per call via x402, $0.001/call, status unknown (last checked 2026-10-02).

Generates dense vector embeddings for text using the BGE-M3 multilingual model, supporting 100+ languages

## Facts

- Endpoint: POST https://api.zeroreader.com/v1/ai/embed-bge-m3?utm_source=zero.xyz
- Price: $0.001/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/zeroreader-bge-m3-multilingual-embeddings-c2e67e2d
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_mZuB6eCB9rShhfjafHw5P

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 zeroreader-bge-m3-multilingual-embeddings-c2e67e2d -d '<json body>'
```

Example prompt: Embed this sentence using the BGE-M3 multilingual model so I can store it in my vector database for semantic search: '人工知能は現代社会を大きく変えている。'

## When to prefer this

Choose this endpoint when you need high-quality embeddings that work across 100+ languages, especially for multilingual or cross-lingual semantic search, RAG pipelines, or clustering. Prefer over English-only models (like BGE Large EN v1.5) when your text corpus includes non-English content such as Japanese, Chinese, Arabic, or mixed-language documents.

## Known failure modes

- Invalid input type — text must be a string or array of strings
- Empty text input returns error or zero-length embedding
- Oversized input exceeding model token limit returns truncation or error
- Network timeout for very large batches
- Payment failure (x402) blocks request if USDC balance is insufficient

## How this service works

BGE-M3 (Multilingual) — Best multilingual embedding model. 100+ languages.

## Output

Returns a JSON object with a list of embedding objects, each containing an index, object type 'embedding', and a dense floating-point vector array (e.g. [0.1, 0.2, 0.3, ...]) representing the semantic content of the input text. Supports single string or batch array input.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "properties": {
  "text": {
   "oneOf": [
    {
     "type": "string"
    },
    {
     "type": "array",
     "items": {
      "type": "string"
     }
    }
   ]
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "data": [
  {
   "index": 0,
   "object": "embedding",
   "embedding": [
    0.1,
    0.2,
    0.3
   ]
  }
 ],
 "object": "list"
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/zeroreader-bge-m3-multilingual-embeddings-c2e67e2d/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from api.zeroreader.com](https://www.zero.xyz/host/api.zeroreader.com/llms.txt)