Chunking
Parse
Split large text documents into smaller, manageable chunks using different chunking strategies optimized for RAG (Retrieval-Augmented Generation) workflows. This endpoint supports multiple chunking algorithms including token-based, sentence-based, recursive, semantic, and specialized strategies.from orq_ai_sdk import Orq
import os
with Orq(
api_key=os.getenv("ORQ_API_KEY", ""),
) as orq:
res = orq.chunking.parse(request={
"text": "The quick brown fox jumps over the lazy dog. This is a sample text that will be chunked into smaller pieces. Each chunk will maintain context while respecting the maximum chunk size.",
"metadata": True,
"strategy": "semantic",
"chunk_size": 256,
"threshold": 0.8,
"embedding_model": "openai/text-embedding-3-small",
"dimensions": 512,
"mode": "window",
"similarity_window": 1,
})
# Handle response
print(res)
import { Orq } from "@orq-ai/node";
const orq = new Orq({
apiKey: process.env["ORQ_API_KEY"] ?? "",
});
async function run() {
const result = await orq.chunking.parse({
text: "The quick brown fox jumps over the lazy dog. This is a sample text that will be chunked into smaller pieces. Each chunk will maintain context while respecting the maximum chunk size.",
metadata: true,
strategy: "semantic",
chunkSize: 256,
threshold: 0.8,
embeddingModel: "openai/text-embedding-3-small",
dimensions: 512,
mode: "window",
similarityWindow: 1,
});
console.log(result);
}
run();
Show Parameters
Show Parameters
{
# one of: TokenChunkerStrategy, SentenceChunkerStrategy, RecursiveChunkerStrategy, SemanticChunkerStrategy, AgenticChunkerStrategy, FastChunkerStrategy, LateChunkerStrategy
"text": str, # required
"metadata": Optional[bool],
"return_type": Optional[Literal["chunks", "texts"]],
"strategy": Union[Union[Union[Union[Union[Union[Literal["token"], Literal["sentence"]], Literal["recursive"]], Literal["semantic"]], Literal["agentic"]], Literal["fast"]], Literal["late"]], # required
"chunk_size": Optional[int],
"chunk_overlap": Optional[int],
"min_sentences_per_chunk": Optional[int],
"separators": List[str], # optional
"min_characters_per_chunk": Optional[int],
"threshold": Union[float, Literal["auto"]], # optional
"embedding_model": str, # required (variant-specific)
"dimensions": Optional[int],
"max_tokens": Optional[int],
"mode": Optional[Literal["window", "sentence"]],
"similarity_window": Optional[int],
"model": str, # required (variant-specific)
"candidate_size": Optional[int],
"system_prompt": Optional[str],
"target_size": Optional[int],
"delimiters": Optional[str],
"pattern": Optional[str],
"prefix": Optional[bool],
"consecutive": Optional[bool],
"forward_fallback": Optional[bool],
}
{
// one of: TokenChunkerStrategy, SentenceChunkerStrategy, RecursiveChunkerStrategy, SemanticChunkerStrategy, AgenticChunkerStrategy, FastChunkerStrategy, LateChunkerStrategy
text: string; // required
metadata?: boolean;
returnType?: "chunks" | "texts";
strategy: "token" | "sentence" | "recursive" | "semantic" | "agentic" | "fast" | "late"; // required
chunkSize?: number;
chunkOverlap?: number;
minSentencesPerChunk?: number;
separators?: string[];
minCharactersPerChunk?: number;
threshold?: number | "auto";
embeddingModel: string; // required (variant-specific)
dimensions?: number;
maxTokens?: number;
mode?: "window" | "sentence";
similarityWindow?: number;
model: string; // required (variant-specific)
candidateSize?: number;
systemPrompt?: string;
targetSize?: number;
delimiters?: string;
pattern?: string;
prefix?: boolean;
consecutive?: boolean;
forwardFallback?: boolean;
}
Show Response
Show Response
{
"chunks": [{
"id": str,
"text": str,
"index": int,
"metadata": { # optional
"start_index": int,
"end_index": int,
"token_count": int,
},
}],
}
{
chunks: {
id: string;
text: string;
index: number;
metadata?: {
startIndex: number;
endIndex: number;
tokenCount: number;
};
}[];
}