SciFigureAI turns text, sketches, and images into scientific figure drafts with PPTX/SVG export. Free plan; paid from $9.90/month billed yearly.
Researched and fact-checked by Bartu Cavusoglu · Updated October 26, 2026
SciFigureAI is an AI scientific figure maker that turns a research idea, paper abstract, or mechanism description into a clean figure draft. The company positions the drafts for papers, posters, grants, and slides, and describes the product as built for fast visual exploration first, followed by researcher review and refinement of the result. Typical outputs include graphical abstracts, mechanism diagrams, and protocol workflows.
The homepage shows example prompts across immunology, cancer biology, cell biology, materials science, chemistry, physics, neuroscience, genomics, microbiology, medical imaging, and ecology. Each example pairs a figure type with a written prompt that users can copy and adapt, which gives new users a concrete starting point instead of a blank canvas. The range suggests the tool is not limited to one discipline.
The workflow has four steps. A user describes the figure by pasting an abstract, mechanism, workflow, or visual direction. SciFigureAI then generates a draft in a project workspace and keeps the result in context. The user edits the selected image with follow-up prompts or uploaded references, and finally exports the result for presentations, manuscripts, and review workflows. A first draft is estimated at 5 credits.
Three input routes are offered: text to figure, sketch to figure, and edit image. Sketch to figure takes a rough uploaded sketch and develops the structure into a more complete scientific visual. Edit image accepts a generated or uploaded image and revises composition and details through follow-up instructions, all within the same project. Previews download free, while file exports use credits.
Core capabilities are text to figure, sketch to figure, edit image, and PPTX or SVG export. The site also lists multiple visual styles, including schematic, 3D, photorealistic, workflow, mechanism, and graphical abstract directions, so users can try different looks for the same concept before settling on one. Everything runs in one project workspace rather than in separate tools.
Export is handled in two tiers. Previews can be downloaded for free, while PPTX and SVG files for slides and later editing cost credits. Having both formats means figures can be adjusted after generation in presentation or design software. Paid plans include commercial use of generated outputs, subject to the Terms and third-party rights, which is worth reading before publishing.
SciFigureAI is an AI scientific figure maker that turns a research idea, abstract, or mechanism description into figure drafts. The drafts are intended for papers, posters, grants, and slides, and can be refined in the same workspace.
A Free plan costs $0 with 5 starter credits. Paid plans are Basic at $9.90 per month, Plus at $29.90 per month, and Enterprise at $59.90 per month, all shown with yearly billing. PAYG credits can also be bought.
Yes. The Free plan includes 5 starter credits for generating scientific figure drafts. The first draft is estimated at 5 credits, so the free credits cover roughly one draft before upgrading or buying PAYG credits.
Basic includes 300 credits per month, about 60 images. Plus includes 800 credits, about 160 images. Enterprise includes 2,000 credits, about 400 images. Monthly credits expire after 30 days.
The site lists PPTX and SVG export. Previews can be downloaded for free, while exporting files for slides and later editing uses credits.
Paid plans and PAYG credits include commercial use of generated outputs, subject to the Terms and third-party rights. Review the Terms of Service before publishing or selling work that uses generated figures.
It suits researchers, students, and labs that need quick first drafts of graphical abstracts, mechanism diagrams, and workflows without design experience. Basic targets occasional use, Plus active paper workflows, and Enterprise labs with high-volume needs.
The sources do not list any third-party integrations. Results move into other tools through PPTX and SVG export.