Text Chunker API — IteraTools

About this tool

POST /text/chunk splits large text into smaller, optionally-overlapping chunks — the first step in any RAG (Retrieval-Augmented Generation) pipeline. Use it to prepare documents for /embeddings and then store them in /memory/upsert.

token — fixed character windows (1 token ≈ 4 chars), reliable for LLM context limits
sentence — groups complete sentences up to chunk_size; preserves readability (default)
paragraph — groups paragraphs separated by double newlines; ideal for structured documents

Max input: 500,000 characters. Overlap prevents context loss at chunk boundaries.

Quick Start

curl -X POST https://api.iteratools.com/text/chunk \
  -H "Authorization: Bearer YOUR_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "text": "Your long document goes here...",
    "chunk_size": 500,
    "overlap": 50,
    "strategy": "sentence"
  }'

Request Parameters

Parameter	Type	Default	Description
`text` *	string	—	Text to chunk (max 500,000 chars)
`chunk_size`	integer	500	Approximate tokens per chunk (1 token ≈ 4 chars)
`overlap`	integer	50	Overlap tokens between consecutive chunks
`strategy`	string	sentence	token \| sentence \| paragraph

Response

{
  "ok": true,
  "data": {
    "chunks": [
      "The quick brown fox jumps over the lazy dog.",
      "This is the second chunk with some overlap context.",
      "..."
    ],
    "count": 12,
    "strategy": "sentence",
    "avg_length": 487
  }
}

RAG Pipeline Example

Use /text/chunk + /embeddings + /memory/upsert for a complete RAG pipeline:

# Step 1: Chunk your document
chunks=$(curl -s -X POST https://api.iteratools.com/text/chunk \
  -H "Authorization: Bearer YOUR_KEY" \
  -d '{"text":"...","strategy":"sentence"}' | jq '.data.chunks')

# Step 2: Embed each chunk
embeddings=$(curl -s -X POST https://api.iteratools.com/embeddings \
  -H "Authorization: Bearer YOUR_KEY" \
  -d "{\"text\": $chunks}")

# Step 3: Store in memory
curl -X POST https://api.iteratools.com/memory/upsert \
  -H "Authorization: Bearer YOUR_KEY" \
  -d '{"namespace":"docs","id":"chunk-1","text":"..."}'

Pricing

$0.001 per request via x402 micropayment on Base (USDC). Zero external API calls — pure local processing.