What you're probably wondering.

A general-purpose LLM reads your work in one pass, with the knowledge it holds at its training cut-off. It does not look for the current literature. It does not verify references. It does not score against field-specific standards. It does not check its own work. It cannot overcome its own bias.

Explore Science does all of these things, in a multi-phase architecture we developed for autonomous scientific research. We orchestrate each frontier model (Claude, GPT, Gemini, Mistral, Grok) with our own Explorer One. We route each sub-task to the model that performs best on it. We verify each citation live to a DOI. We hold your manuscript in context across hours of analysis, not one two-minute pass. The depth is the difference. You see it in the nuance and insight of our feedback.

As long as the work demands, often upwards of an hour, depending on the paper and how much you ask of it. In that time it reads your whole manuscript across multiple models, assembles the current literature, calculates a calibre score, and verifies every reference to a live DOI. It runs until the analysis is genuinely done, not on a clock, and that same depth applies whether you're running a full review, a novelty check, or any of the other tools.

You don't have to watch it run; we'll let you know the moment it's ready.

What you get back isn't a verdict, as whether a paper is ready is a judgment that belongs to you, and ultimately the editor. You should be wary of any tool that claims it can tell you your paper is ready. What Explore Science does is give you everything you need to make the call with your eyes open: it scores your manuscript against field-specific standards and surfaces the issues most likely to get a paper desk-rejected, from claims the evidence doesn't support to methodological gaps, scope or journal-fit problems, and conclusions that overreach. Each issue comes with a suggested change before an editor sees it, and you can inspect how the judgment is made on the scoring methodology page. Our aim is to help you send the strongest version of the paper the first time.

Every review produces a Calibre score: a single number out of 100. Eight dimensions of the work each carry their own sub-score, from research design and statistical rigour to how well the evidence supports the conclusions. It is not a readability grade. The score combines the quality of the work with the evidential weight of its study design. A well-written paper on thin evidence therefore scores honestly, not generously. Read more about the methodology here. The goal is an honest read of the paper, and a score to revise against.

Every reference is validated against several academic databases on the basis of persistent identifiers (a DOI where one exists) and triangulation of core fields where one doesn't. If a reference doesn't resolve, it's discarded before it ever reaches your review or your literature search. You can click any citation in any output and land on the actual paper.

A mixture. The system selects it task by task. Models differ by sub-task, by reviewer role, and by scientific field. The orchestrator routes each step to the model that fits it best.

The current mixture includes (but is not limited to) Claude, ChatGPT, Gemini, Mistral, and Grok, with our own in-house Explorer One. You do not select a model. You get the strongest answer at each step: a consensus across models that cross-checks and removes single-model bias.

No. We do not train models on user manuscripts. Your work is yours.

We have coverage across all fields, including the life sciences, medicine, computer science, psychology, physics, economics, and the social sciences. Whatever your manuscript's field, the system automatically finds and assembles the relevant, up-to-date literature to work from.

Across the dozens of subfields we've reviewed, it performs best on evidence-driven research where conclusions follow from data and method. If your paper sits in an unusually specialised subfield and you’d like to sanity-check fit, contact us before purchasing.

Explore Science does two kinds of work: evaluative (it reviews, scores, and critiques your manuscript) and generative (coming soon; as a co-scientist, it helps you develop ideas, study designs, and analysis). In both cases you direct the work and remain the sole author, accountable for everything you submit.

Evaluative use is equivalent to seeking pre-submission feedback from a colleague, which is standard academic practice; most journals allow or encourage it. Generative use is like working with any research assistant or tool: it does not earn authorship, and most journals now ask you to disclose substantive AI assistance.

Policies vary, so when in doubt, acknowledge how you used Explore Science in the submission process, as you would acknowledge any substantive reviewer or tool.

90% of users rate a review as equal to or better than human peer review. From 304 head-to-head ratings by our own users, self-reported.

Human peer review is unpaid and often rushed. At times the reviewer can be a non-specialist in your exact topic. Explore Science gives consistent rigour, subject-matter depth, and a careful read to each submission. You get it in hours, not months.

The goal is that your paper reaches a human reviewer in the strongest form you can send.

Yes. Start your first project free, no card required, and see the quality of our tools before deciding to pay - the single-project free plan lasts as long as you need. If you want premium features or more usage allowance, paid Researcher plans start at $100/month ($50 with student pricing), with Pro at $200/month, differing only in usage quota. A one-off paper review with no subscription is $49. Full plan details are on the pricing page.

Principal Investigators, lab and department heads, CROs, R&D teams, and research collaborations can contact us about our lab and institutional plans, which offer linked billing, multiple seats, SSO, and a security review.

We also run contract research, where our autonomous system runs original research against a dataset you already hold. Talk to the lab to learn more.

Paper-Wizard brand has been unified under Explore Science, the lab which built it. The product, the team, the engine: all the same, and always improving. Your account, projects, and reviews carry over automatically, so existing users land right where they left off. Everything that worked before still works, you just have access to many more tools to support your research and scientific workflows.