Dashboard
Sparrow usage patterns — processing volume, model distribution, latency, and geographical reach.
Total Inferences
—
Success Rate
—
Avg. Duration
—
per inference
Top Model
—
Avg Duration by Page Count
Average seconds per inference, grouped by page count
No data
Inference by Model
Share of inferences by Vision LLM model
No data
Inference Events
Duration vs time — bubble size encodes page count
No events to display
Inference Requests by Country
Distribution of total calls
No data
Unique Users by Country
Distribution by IP address
No data
Document Extraction
Structured data from invoices, receipts, statements, and tables — multi-page PDF, page classification, bounding-box annotation, and schema validation. No cloud dependencies.
Business Rules
Push formatting, derived fields, classification, and transformations to the LLM itself — no post-processing code, just typed schema fields with optional defaults.
Sparrow Agent
Orchestrate Vision LLM extraction with Text LLM reasoning — chain classification, extraction, and field validation with visual monitoring and error handling.