Pitch your Startup, App or Hardware or post a Startup Event or Startup Job
Company profile:
Startup name: PaperFig
Tagline: AI scientific figures with editable labels, built-in OCR, and 4K
Elevator Pitch: AI scientific figure maker for researchers—turn ideas, sketches, references, or PDFs into scientific figure drafts with OCR-assisted label editing.
Unlike general AI image generators, PaperFig is built specifically for researchers: it turns a text brief, sketch, reference image, or PDF into a scientific figure draft, and its built-in OCR lets you review and correct labels without rebuilding the figure from scratch.
Target Market: Researchers and academics who need to create scientific figures, mechanism diagrams, workflows, and graphical abstracts for papers, grants, and presentations.
How will you make money?: Freemium. Free plan with signup credits; paid plans from $16/month or $144/year (Scholar, Pro Scholar, Lab Pro).
How much capital have you raised?: None
Website: https://paperfig.io
City/Country:
AI-assisted summary:
PaperFig is a web-based AI scientific-figure generator aimed at researchers who need to turn written descriptions, rough sketches, photos, PDFs, or reference images into editable scientific illustrations. (Source: https://paperfig.io/) Its relevance lies in targeting a narrow but time-sensitive research workflow: preparing mechanism diagrams, workflows, graphical abstracts, and other visuals for manuscripts, presentations, and lab meetings. (Source: https://paperfig.io/)
Problem and target users
Scientific researchers often need clear visual explanations of methods, mechanisms, pathways, and results, but producing those figures can require illustration skills, manual layout work, and repeated revisions. PaperFig positions its product for researchers across biomedical science, chemistry, materials science, engineering, environmental science, computer science, and related fields. (Source: https://paperfig.io/)
The target user appears to be an individual researcher or lab preparing papers, grant materials, presentations, teaching content, or revisions, rather than a general-purpose design customer. (Source: https://paperfig.io/) This is also a category where accuracy and disclosure matter: Springer Nature states that AI-assisted visual content must be transparent, based on verifiable source material where applicable, reviewed by humans, and accurately represent the underlying evidence. (Source: https://www.springernature.com/gp/policies/editorial-policies/ai-manuscript-preparation)
The Royal Society of Chemistry similarly says authors remain responsible for AI-produced content, must disclose relevant generative-AI use, and should check the accuracy of AI-created graphics. (Source: https://www.rsc.org/publishing/journals/processes-and-policies/author-responsibilities)
Product and solution
PaperFig’s workflow starts with user-supplied material: a paragraph from a paper, a sketch, an existing figure, a lab photo, a PDF, or a reference image. (Source: https://paperfig.io/) Users then specify what the figure should show, including layout, labels, and journal style, before generating a draft for revision and export. (Source: https://paperfig.io/)
- PaperFig supports text-to-figure, sketch-to-figure, photo-to-figure, PDF-based, and reference-image inputs. (Source: https://paperfig.io/)
- The product says it can be used for mechanism diagrams, pathway illustrations, experimental workflows, conceptual frameworks, flow charts, graphical abstracts, and related scientific schematics. (Source: https://paperfig.io/)
- Its built-in OCR-based editor is intended to let users modify labels, annotations, and captions without regenerating an entire figure. (Source: https://paperfig.io/)
- Available export options include PNG, a 4K PNG upgrade, and editable SVG and PPTX label exports after label correction. (Source: https://paperfig.io/)
PaperFig explicitly acknowledges a core limitation of generative tools: AI-created labels can be wrong, and the company instructs users to check and correct them before export. (Source: https://paperfig.io/) Its terms also state that the service may process prompts, uploads, PDFs, sketches, and reference materials through third-party AI and other service providers. (Source: https://paperfig.io/terms)
Business model, pricing signals, and traction
PaperFig uses a self-serve, credit-based SaaS model with a free tier, annual subscriptions, and add-on credit packs. (Source: https://paperfig.io/pricing) The free plan includes 150 signup credits, text-to-figure generation, label editing, and watermarked PNG downloads for non-commercial use; each figure is listed as costing roughly 50 credits. (Source: https://paperfig.io/pricing)
Annual plans listed on PaperFig’s pricing page include Scholar at $144 per year, Pro Scholar at $240 per year, and Lab Pro at $648 per year. (Source: https://paperfig.io/pricing) Paid plans add watermark-free downloads, publication and commercial licensing, TIFF/PDF export, and editable SVG/PPTX exports. (Source: https://paperfig.io/publication-license)
For users handling unpublished or sensitive research material, PaperFig advises against uploading confidential patient data, proprietary datasets, or restricted institutional materials unless organizational policy permits use of AI-assisted tools. (Source: https://paperfig.io/)
Expert take
PaperFig’s strongest positioning is its focus on research communication rather than generic image generation: structured scientific figures, editable labels, export formats, and publication-oriented licensing address a specific researcher workflow.
The central execution risk is trust. In scientific publishing, visual polish is not enough; users must be able to validate every label, relationship, and implied scientific claim, while also following each target journal’s AI-use and disclosure rules.
For founders and investors, the key question is whether PaperFig can become embedded in repeat lab and manuscript workflows rather than remaining a one-off figure-generation tool. Its credit-based model and lab tier create a plausible path to recurring usage, but public product claims should be evaluated alongside demonstrated scientific accuracy and customer adoption.
Note: Information based on publicly available sources at the time of writing, and summarized by AI.
Pitch your Startup, App or Hardware or post a Startup Event or Startup Job
About the Author
