AI / LLM
RAG Chunk Planner
Plan chunk sizes and overlap for better RAG.
Document Input
Chunking Strategy
Fixed: slides a window of exact token size with overlap — best for embedding models with hard limits.
Chunking Parameters
Embedding Model & Cost
Results
—
Chunks
—
Tokens / chunk
—
Overlap tokens
—
Total tokens to embed
—
Est. embedding cost
stride = chunkSize − overlap |
chunks = ⌈(totalTokens − overlap) / stride⌉ |
totalToEmbed = chunks × chunkSize
Coverage Visualisation
Enter parameters to see coverage.
Unique content
Overlap region
Chunk boundary
Config Exports
Export your RAG configuration for use in your pipeline.
Saved Workspaces