# ONE Engine Word Bigrams (T5) Tokenizer

> ONE Engine Word Bigrams (T5) Tokenizer is a paid API for AI agents from one-search.one-engine.workers.dev, paid per call via x402, $0.0025/call, status unknown (last checked 2026-10-02).

Extracts word-level bigrams from input text using T5 tokenization, returning token pair sequences for NLP analysis.

## Facts

- Endpoint: POST https://one-search.one-engine.workers.dev/v3/text-token/word-bigrams/t5?utm_source=zero.xyz
- Price: $0.0025/call
- Payment: x402
- Status: unknown
- Last checked: 2026-10-02
- Activations on Zero: 0
- Tags: x402
- Canonical page: https://www.zero.xyz/c/one-engine-word-bigrams-t5-tokenizer-82b31360
- Structured record (JSON): https://api.zero.xyz/v1/capabilities/cap_g5f4-aiU3dYwcmpa6zTZP

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 one-engine-word-bigrams-t5-tokenizer-82b31360 -d '<json body>'
```

Example prompt: Extract all word bigrams from this text using T5 tokenization: 'The quick brown fox jumps over the lazy dog. Natural language processing enables machines to understand human text.'

## When to prefer this

Choose this endpoint when you specifically need T5-tokenized word bigrams from text as part of an NLP pipeline, feature extraction workflow, or linguistic analysis — particularly when you want a low-cost, pay-per-call microservice on Base without managing infrastructure. Prefer this over general NLP libraries when building autonomous agents that need serverless, metered bigram extraction.

## Known failure modes

- Text exceeding 50,000 characters may be rejected or truncated
- Empty or whitespace-only text may return empty bigram arrays
- Payment failure via x402/Base USDC may block the request entirely
- Very short texts (single word) produce no bigrams
- Non-Latin scripts or unusual Unicode may produce unexpected tokenization results

## How this service works

Low-cost pay-per-call utility APIs for autonomous agents using x402 on Base.

## Output

Returns a JSON object containing word-level bigrams derived from T5 tokenization of the input text — likely including arrays of two-token word pairs or sequences extracted across the input, suitable for downstream NLP tasks such as feature engineering, language modeling, or linguistic analysis.

## Request schema (JSON Schema)

```json
{
 "type": "object",
 "required": [
  "text"
 ],
 "properties": {
  "text": {
   "type": "string",
   "description": "Text up to 50000 characters"
  }
 }
}
```

## Response schema (JSON Schema)

```json
{
 "type": "object",
 "additionalProperties": true
}
```

## More

- Live health (JSON, refreshed every minute): https://www.zero.xyz/c/one-engine-word-bigrams-t5-tokenizer-82b31360/health.json
- [Zero catalog index](https://www.zero.xyz/llms.txt)
- [Other services from one-search.one-engine.workers.dev](https://www.zero.xyz/host/one-search.one-engine.workers.dev/llms.txt)