Shared timeline · September 2023–August 2026

From 2023 to 2026, the test expanded from isolated charts to languages, documents, and dashboard state.

Model and harness releases expanded what a general system could see and do. Research systems added structured plans, direct editing, compilers, and feedback. Benchmarks added real data, multi-turn refinement, professional domains, deception, chart pairs, multi-chart figures, multilingual reading, document retrieval, browser behavior, and analytical use.

How to read this timeline. Horizontal position marks the release or publication date; label width does not encode duration. Read across to compare changing scope, not to infer that a platform release caused a research result. Publication dates can also lag the work.
Model and harness Research system Benchmark
01

Model and harness baseline

What a general agent can see, execute, and reuse

  1. GPT-4V adds image input

    General models can inspect visual artifacts.

  2. GPT-4o · computer use

    Native multimodality, then screen, cursor, click, and typing actions.

  3. coding agent · agent tools · GPT-5

    Terminal coding, tools, tracing, and stronger agentic coding enter normal harnesses.

  4. Hosted execution and reusable skills

    Shell, persistent workspaces, reusable skills, and compaction become platform primitives.

02

Research systems and techniques

How systems structure authoring and correction

  1. Data Formulator 2 · NL4DV-LLM

    Visible transformed data, direct controls, history, and explicit analytic specifications.

  2. PlotGen · DashChat · nvAgent

    Multimodal feedback, dashboard languages, schema planning, execution, and repair.

  3. NL2Dashboard · Raiven

    Compact dashboard plans with atomic edits; restricted scientific language with deterministic compilation.

03

Benchmarks and measured capability

What the field can test repeatably

  1. Plot2Code · VisEval

    Plot reproduction and database visualization, with execution and visual checks.

  2. Text2Vis

    1,985 tasks join data, questions, answers, code, and annotated charts; targeted review improves pass rate.

  3. RealChart2Code · Chartography · ChartDiff · PolyChartQA · Chart-MRAG · DashArena

    Real data, professional reading, chart pairs, multi-chart and multilingual QA, document retrieval, state, and browser replay.