All tools

Chunk Text

AI / Agent

Split documents into overlapping chunks for RAG pipelines: by sentence, paragraph, or fixed size with configurable overlap.

Getting chunk size wrong is one of the more common ways a RAG pipeline underperforms, so this splits a document by paragraph or fixed size with a configurable overlap percentage, so adjacent chunks share enough context that a fact doesn't get sliced across a boundary. Sentence-based chunking is the better choice for shorter, semantically coherent content like FAQ or support articles rather than long-form prose. A reasonable starting point is 200-500 tokens with 10-20% overlap, though the right numbers depend on your embedding model's limits and how granular your retrieval needs to be. This does structural chunking only, by sentence, paragraph, or size, not semantic chunking based on meaning, which would need an embedding- or LLM-based approach this tool doesn't attempt. No LLM call happens here; splitting is plain client-side text logic.

chunkragsplitoverlapretrievalembeddings

How to use Chunk Text

  • 1.Paste a long document and choose paragraph or fixed-size chunking to split it into pieces sized for your RAG pipeline's embedding step.
  • 2.Set an overlap percentage so adjacent chunks share context, reducing the chance that a fact gets cut across a chunk boundary.
  • 3.Use sentence-based chunking for shorter, semantically coherent chunks when you're indexing FAQ or support content rather than long-form prose.

Frequently asked questions

Does this run in my browser or on a server?
Chunking runs entirely client-side with simple text-splitting logic — no LLM call is involved, and your document isn't uploaded.
What chunk size should I use?
It depends on your embedding model's context limit and retrieval granularity, but 200-500 tokens with 10-20% overlap is a common starting point for RAG.
Does it split on semantic meaning?
No — this does structural chunking (by sentence, paragraph, or fixed size), not semantic chunking. For meaning-aware splits, you'd need an embedding-based or LLM-based approach.
Can I chunk very large files?
The browser tool handles typical document sizes fine; for very large corpora processed in bulk, use the REST API or SDK to script the chunking step.

Use via API, SDK, or MCP

cURL# Free: 1,000 req/day · Pro: 10,000 req/day
curl -X POST https://api.utilix.tech/v1/tools/chunk-text \
  -H "Authorization: Bearer utx_live_..." \
  -H "Content-Type: application/json" \
  -d '{"text":"Long document text goes here. Split into chunks for embedding and retrieval.","chunkSize":200,"overlap":20,"strategy":"paragraph"}'

Get an API key from your dashboard · Full API docs →

Related tools

Query ExpansionExpand a search query with synonyms and stemmed variants to improve retrieval recall in RAG pipelines.AI / AgentRelevance ScoreScore a single text passage against a query. Returns a 0-1 relevance score, grade, matched terms, and coverage.AI / AgentRerank ChunksScore and rerank text chunks by relevance to a query using TF-IDF. Drop irrelevant chunks before LLM context.AI / Agent