KXAI

KXAI: a research guide grounded in the papers you're actually reading.

Not a general chatbot. KXAI answers from the paper you have open, right now, architecture, results, gaps, equations, all cited back to the tab it came from. Open it from the button in the corner of any page, or try it below.

Grounded in papers below — never invented

I'm KXAI. Ask me about papers, architectures, results, or how KinetixAI works. I'll cite my sources.

Demo mode — answers come from the papers in your catalog.

What it actually does

01

Architecture, broken down

Ask what a paper's system looks like and get the real components, pulled from its Architecture tab, not a generic summary.

02

Results you can check

Ask for the key numbers and get the actual benchmark results, cited back to the Cheatsheet tab.

03

Gaps and what's next

Ask what's missing or what to read next and get the paper's own stated limitations, plus the paper it leads to.

04

A mnemonic that sticks

Ask for a way to remember it and get the paper's hook, built to actually stay with you.

05

Equations and terms, on demand

Ask for the math or a definition and get it straight from the Cheatsheet, no hunting through the PDF.

06

Cross-paper search

Ask which paper in your bucket covers a topic, and KXAI searches all of them, then offers a one-click switch.

Who it's for

Students

Ask the question you'd be embarrassed to ask in office hours, and get a cited answer from the actual paper instead of a half-remembered explanation.

Teachers

Point a class at KXAI instead of forwarding a PDF with no way to check a claim, every answer cites the tab it came from.

Researchers

Check whether your read of a paper's architecture or results is right, in seconds, cited back to the source, before it goes in a lit review.

Professionals

Get the one detail you need before you ship, the actual mechanism or the actual number, not a confident guess.

PhD students

Keep a whole literature review honest. Check a claim against the actual paper before it goes in a chapter, not after.

Labs & teams

Point a whole lab or class at the same grounded assistant instead of everyone re-reading the same paper alone. Pairs with Campus and Teams plans.

Honest about where this is

What's real today, and what's coming

KXAI runs in demo mode: every answer above is grounded in content we wrote for the papers in your bucket (Attention Is All You Need is free to try now, three more are in Paper Lab). Mapping an arbitrary paper of your own, the real headline capability, needs a backend that can read and understand any paper. That isn't live yet, so asking for it captures the request honestly, an email so we can notify you, instead of faking a result. The same box starts generating for real the day it ships, no rebuild needed.

What's free, what's paid

KXAI's demo-mode Q&A is free on Explorer. Unlimited use across your whole bucket, plus your own Research RAG, is a Scholar feature.

Explorer

$0forever

Taste the whole platform.

Create account

Explore all hubs
A few visualizations a day
Ask KinetixAI, Story tier
Community support