mirror of
https://github.com/Significant-Gravitas/AutoGPT.git
synced 2026-04-08 03:00:28 -04:00
Update block docs for: jina/chunking.md
This commit is contained in:
30
docs/content/platform/blocks/update/jina/chunking.md
Normal file
30
docs/content/platform/blocks/update/jina/chunking.md
Normal file
@@ -0,0 +1,30 @@
|
||||
|
||||
## Jina Text Chunking
|
||||
|
||||
### What it is
|
||||
A specialized tool that breaks down large texts into smaller, manageable pieces using Jina AI's advanced text segmentation service.
|
||||
|
||||
### What it does
|
||||
This component takes long texts and intelligently divides them into smaller chunks while maintaining context and meaning. It can also provide detailed information about how the text was divided if needed.
|
||||
|
||||
### How it works
|
||||
1. The tool connects to Jina AI's segmentation service using your security credentials
|
||||
2. It processes each text you provide, splitting it according to your specifications
|
||||
3. It returns the smaller chunks and, if requested, information about how the text was divided
|
||||
4. All chunks maintain context and are sized according to your requirements
|
||||
|
||||
### Inputs
|
||||
- Texts: A collection of text documents you want to divide into smaller pieces
|
||||
- Credentials: Your Jina AI security credentials for accessing the service
|
||||
- Maximum Chunk Length: The largest size you want for each chunk (default is 1000 characters)
|
||||
- Return Tokens: Option to receive detailed information about how the text was divided (default is No)
|
||||
|
||||
### Outputs
|
||||
- Chunks: The collection of smaller text segments created from your original texts
|
||||
- Tokens: (Optional) Detailed information about how each piece of text was divided
|
||||
|
||||
### Possible use cases
|
||||
- Breaking down large documents like research papers into more manageable sections
|
||||
- Preparing content for AI analysis systems that work better with smaller text segments
|
||||
- Creating summaries or excerpts from longer documents
|
||||
- Processing large amounts of text data for content management systems
|
||||
Reference in New Issue
Block a user