đ Dark Mode
đ Chonkie Text Chunker - Enhanced
Advanced text chunking with full Chonkie library support
đ Input & Configuration
âī¸ Text Input
đ File Upload
Text to chunk:
This is a sample document for testing advanced text chunking capabilities. It contains multiple sentences across several paragraphs. The Chonkie library will split this text into manageable chunks based on the selected strategy. Each chunk will maintain semantic coherence while respecting the specified size limits. Advanced chunkers can use neural networks or large language models for more sophisticated splitting strategies.
Upload document:
Override character limit
â ī¸ (May cause performance issues)
Chunker Type:
Token Chunker
Sentence Chunker
Recursive Chunker
Semantic Chunker
EMBEDDINGS
Code Chunker
Neural Chunker
AI
Late Chunker
EMBEDDINGS
Slumber Chunker
LLM
Splits text into chunks based on sentences
Tokenizer Type:
Character Tokenizer
Word Tokenizer
Chunk Size:
100
Chunk Overlap:
20
âī¸ Advanced Options
Embedding Provider:
Sentence Transformers (Local)
OpenAI Embeddings
Cohere Embeddings
Google Gemini
Jina AI
Voyage AI
Embedding Model:
all-MiniLM-L6-v2
all-mpnet-base-v2
multi-qa-MiniLM-L6-cos-v1
Semantic Threshold:
0.3
Lower values create more chunks, higher values create fewer chunks
Similarity Window:
3
Number of sentences to compare for similarity
Min Sentences per Chunk:
1
Minimum sentences required in each chunk
â ī¸ (Beta - may not function)
Min Characters per Sentence:
24
Minimum sentence length to be considered valid
â ī¸ (Beta - may not function)
Include Delimiter:
Previous (attach to preceding chunk)
Next (attach to following chunk)
None (exclude delimiters)
How to handle sentence delimiters in chunks
Language (optional):
Auto-detect
English
Spanish
French
German
Italian
Portuguese
Russian
Chinese
Japanese
đž Configuration Management
Select a saved configuration...
đĨ Load
đž Save
đī¸ Delete
Save your current settings as a reusable configuration for batch processing.
Processing chunks...
đ Analyze & Chunk Text
âšī¸ Cancel
đ Results
Configure your settings and click "Analyze & Chunk Text" to see results...