SciFig

SciFig is an AI-powered web application that generates scientific figures from text, sketches, PDFs, photos, or reference images, producing editable, publication-ready outputs for researchers.

What is SciFig?

SciFig is a free online AI scientific figure generator that converts text descriptions, sketches, PDFs, photos, or reference images into journal-ready figures. It supports six input modes and six publication styles, including scientific, line, 3D, editorial, sketch, and watercolor. Outputs are editable: users can change text labels, fonts, and sizes directly, and export to PNG, JPG, PPTX, or layered SVG. The platform is built for researchers at institutions like Stanford, MIT, and Yale.

Key Features

Six input modes — Text, sketch, PDF, photo, reference, and vector canvas; accept any research-related input. Six publication styles — Scientific, line, 3D, editorial, sketch, watercolor, plus auto mode that follows the prompt. Editable outputs — All text and elements remain editable after generation; built-in vector canvas for element-level edits. High-resolution exports — PNG/JPG up to 8K resolution, meeting 300 DPI for journal single-column figures at 1K. Multiple export formats — Editable PPTX, layered SVG, and standard PNG/JPG; no Illustrator roundtrip needed. Multimodal enhance — Upscale figures and refine with additional context from PDFs or images.

Who is it for?

Researchers — Generate mechanism diagrams, signaling pathways, and experimental workflows for journal submissions. PhD students and postdocs — Create clear figures for papers and theses without design experience. Lab PIs and research teams — Standardize figure quality across lab members; review and approve AI-generated drafts. Science educators and communicators — Produce illustrations for lectures, posters, and public outreach.

What can you do with SciFig?

Journal submissions — Generate a figure from a text prompt or a PDF, edit labels, export at 8K PNG at 300 DPI, and submit to journals like Nature, Cell, or Science. Grant proposals — Convert a 40-page grant PDF into presentation-ready figures in under an hour using PDF-to-Figure mode. Posters and slides — Create figures in 16:9 PPTX format with editable shapes for conferences and talks.

How does SciFig work?

1. Start with any input — Type a prompt, upload a sketch, PDF, photo, or reference. SciFig generates a draft in seconds. 2. Refine without starting over — Click any text label to edit it, open the canvas for element-level changes, or attach a PDF for context. Upscale to 8K if needed. 3. Deliver and export — Export to editable PPTX, layered SVG, or PNG/JPG. All elements stay editable for later modifications.

FAQ

Is SciFig free to use?

Yes, SciFig offers a free tier with 100 credits on signup (up to 300 with an invite). Free credits regenerate figures but are for non-commercial evaluation only.

What export formats does SciFig support?

SciFig exports PNG (up to 8K), JPG, PPTX (editable slide), and layered SVG (via vector canvas). PPTX and SVG retain editable shapes and labels.

Does SciFig meet journal resolution requirements?

Yes. At 1K resolution, figures exceed 300 DPI at single-column width; 2K covers double-column figures. Upscale to 4K/8K for posters and covers.

Who owns the copyright of figures created with SciFig?

Paid plan users own full copyright and can publish in journals, thesis, and commercial products. Free plan users have non-commercial rights only.

Can I edit labels after generation?

Yes. Every text label is directly clickable and editable. The built-in vector canvas allows further editing of shapes and elements with undo/redo.