Skip to main content

​
How RAG Works

StepProcess
1. UploadAdd files or connect external sources
2. ChunkDocuments split into ~250 token segments
3. EmbedVector embeddings generated for each chunk
4. SearchUser query matched against embeddings
5. RetrieveTop chunks injected into LLM context

​
Create Dataset

curl -X POST https://api.cuadra.ai/v1/datasets \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: create-ds-001" \
  -d '{"name": "Product Docs", "description": "API guides"}'

​
Upload Documents

Adding documents is a two-step process: upload the file, then associate it with a dataset.

​
Step 1: Upload File

curl -X POST https://api.cuadra.ai/v1/files \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Idempotency-Key: upload-001" \
  -F "file=@product-guide.pdf"

​
Step 2: Associate with Dataset

curl -X POST https://api.cuadra.ai/v1/files/file_abc123/associations \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"datasetId": "ds_xyz789"}'

​
Supported Formats

FormatExtensionsMax Size
PDF.pdf50MB
Word.docx50MB
Text.txt, .md50MB
Data.csv, .json50MB

​
External Connectors

Sync content from external data sources.
ConnectorStatus
Google Drive✅ Available
Notion✅ Available

​
Connect Google Drive

curl -X POST https://api.cuadra.ai/v1/connections \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{
    "connectorSlug": "google_drive",
    "redirectUrl": "https://your-app.com/oauth/callback"
  }'
Connect a dataset to enable RAG: 
curl -X POST https://api.cuadra.ai/v1/models/model_abc/datasets \
  -H "Authorization: Bearer YOUR_TOKEN" \
  -H "Content-Type: application/json" \
  -d '{"datasetId": "ds_xyz", "usageType": "rag"}'

​
Best Practices

DoAvoid
Clean formatting before uploadScanned images without OCR
Use descriptive filenamesDuplicate content across files
Split large docs into sectionsMixing unrelated topics
Group related contentPII or sensitive data

​
Specifications

SpecValue
Max file size50MB
Chunk size~250 tokens
Search latency40-120ms

​
FAQ

​
What file formats work best?

Markdown and plain text yield the best results. PDFs work well if they’re text-based (not scanned images). Use OCR preprocessing for scanned documents.

​
How often is content re-indexed?

Uploaded files are indexed once at upload. External connectors (Google Drive) sync based on your configuration—typically every 1-24 hours.

​
Can I preview what chunks were created?

Not via API currently. Use the Dashboard → Datasets → View to inspect chunks.

​
How do I improve retrieval quality?

  1. Use specific, descriptive filenames
  2. Add summaries at the start of documents
  3. Remove boilerplate/headers that repeat across pages
  4. Split very long documents into logical sections

​
What happens if I delete a document?

The document and its chunks are removed. This affects new chats only—existing chat histories retain their context.

Chat API

Use RAG in chat completions

Models

Link datasets to models