Visual Lab

The Visual Lab plans and renders scientific figures from a brief, then applies a review pass for hierarchy, labels and connections before the researcher approves the output.

Check availability Explore the workspace
SixSentences_· Research, on the recordIn plain termsp. 02

In plain terms

  • 01

    Ground the brief

    Use a dataset, literature search, library source or explicit prompt as the material from which the figure is planned.

  • 02

    Control the output

    Choose figure type, canvas ratio, output resolution and review depth before rendering.

  • 03

    Redraw a real sketch

    Upload a whiteboard photo or rough diagram and preserve its labels and structure in a cleaner visual treatment.

SixSentences_· Research, on the recordA documented workflowp. 03
A documented workflow

01 / 04

Describe the scientific job

State what the reader must understand and attach the relevant data, source or sketch.

02 / 04

Plan content and style

The workflow separates what belongs in the figure from how it should be visually organised.

03 / 04

Render and review

Generate the selected canvas, then check hierarchy, labels, connections and scientific consistency.

04 / 04

Approve and export

Inspect every number and word, save source figures where permitted and export at the required resolution.
Visual Lab · demo workspace01 / 04
SixSentences_ Visual Lab with figure controls and a gallery of publication-ready scientific visuals
Fig. 1The Visual Lab of the demo workspace: figure types, output size, canvas and review passes above, and four finished visuals below.
Fig. 1The Visual Lab of the demo workspace: figure types, output size, canvas and review passes above, and four finished visuals below.
SixSentences_· Research, on the recordMethod and boundariesp. 04

Method and boundaries

01 / 02

Scientific diagrams often fail because content planning and image rendering are collapsed into one step. The Visual Lab treats reference retrieval, content selection, style planning, rendering and review as separate parts of the workflow.

That structure is inspired by research on agentic scientific illustration, including PaperBanana. SixSentences_ adds project grounding and explicit output controls so the visual can be connected to the evidence or data it represents.

  • Method, architecture, flow, concept and data-plot modes
  • 1K, 2K and 4K output options
  • Multiple canvas ratios and review passes
  • Grounding in study data, searches and library sources

02 / 02

A polished image can still contain a wrong label or misleading relationship. The interface therefore makes the review stage visible and reminds authors to approve every word, number and connection before manuscript use.

Figures found inside a paper can be saved with source information when reuse is appropriate. Copyright and licence terms still determine whether a figure may be republished.

SixSentences_· Research, on the recordQuestions, answered.p. 05

Visual Lab

  1. 01Can I choose the output size?

    Yes. The figure composer exposes output resolution and canvas ratio before rendering.

  2. 02Can the figure use my dataset?

    A figure can be grounded in a connected dataset or analysis result so the visual workflow starts from project material.

  3. 03Are generated figures scientifically verified?

    The workflow includes review passes, but authors must still verify every label, number and relationship before publication.

SixSentences_· Research, on the recordContinue through the workspacep. 06

Visual Lab

Fig. 2The seven workflows, this one lit. Each line joins a guide to one it continues to; hover a workflow to follow its lines.
Check availability Explore the workspace

hello@sixsentences.com Open source on GitHub