Agentic AI Reference Architecture

Knowledge & Data · Software component

Chart Data Extractor

Software componentKnowledge & DataKnowledge & Dataarc:ChartDataExtractor

An image-to-text grounder that converts charts, plots, graphs and tables into linearized tables preserving exact values and structure.

Responsibility. Extracts structured numeric data from chart images as linearized tables.

Also known as: Chart-to-table linearizer, Chart linearization

Variant of Image-to-Text Grounder abstract

When to choose. Choose for information-dense images (charts, plots, graphs, tables) where users need exact figures and computation, e.g., financial or benchmark charts.

sends data toinvokesis invoked byis routed to byalternative tospecializesAgent Controller: sends data toAgent ControllerInference Server: invokesInference ServerDocument Ingestor: is invoked byDocument IngestorMultimodal Content Router: is routed to byMultimodal Content RouterImage Captioner: alternative toImage CaptionerImage-to-Text Grounder: specializesImage-to-Text Grounder
Direct neighbourhood (hover for relationship types)

Relationships

invokes dependency

is invoked by dependency

is routed to by dynamic

sends data to dynamic

alternative to variability

Design guidance

Classification

Patterns
Plot-to-table translationTreat charts as structured data
Technologies
DePlotNVIDIA Neva
Quality attributes
Functional suitability: correctness and validity (ISO/IEC 25010 | NIST AI RMF: valid)Maintainability (ISO/IEC 25010)
Risks mitigated
Loss of exact figures in natural-language descriptionsText-only pipelines losing precise chart values (alt-text or skipped charts)

Sources

  1. Ch2.7: T. Nguyen, "Multimodal RAG Approaches," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 2.7. ISBN: 9798244538229.
  2. Ch7.5: T. Nguyen, "NeMo Curator, Riva Speech AI & Multimodal Integration," in Mastering Agentic AI Systems: Guide for the NVIDIA NCP-AAI Exam, 1st ed. 2026, ch. 7.5. ISBN: 9798244538229.
  3. Ref2.07: NVIDIA Developer, "Building multimodal AI RAG with LlamaIndex, NVIDIA NIM, and Milvus | LLM app development," YouTube. Accessed: Sep. 26, 2026. [Online Video]. Available: https://www.youtube.com/watch?v=NaT5Eo97_I0