Acavya

How does Acavya run a coaching test series with CBT, OMR and ranks?

Acavya builds tests in the JEE Main, JEE Advanced, NEET, CUET, Olympiad or a custom pattern, runs them as an online CBT that resumes with the remaining time after a power or internet cut or as phone-scanned OMR, ranks with your tie-breaks, re-ranks after a bonus or changed key and sends a report to parents on WhatsApp.

Last reviewed:

Who is it for, and what hurts today?

  • Test coordinator: typesetting, printing, an OMR scanner, ranks in Excel and re-ranking by hand after a bonus question.
  • Student: a power cut in the middle of a CBT that loses the paper. Answers are cached locally and the student resumes with the remaining time.
  • Faculty: DPPs and papers made by hand, and tests that AI could shortcut.
  • Parent: a rank card days later, with no chapter-wise view.
  • Centre owner: ranks leaked into public lists. Results are private by default.

What can it do?

  • Available now: Patterns for JEE Main, JEE Advanced with partial marking, NEET, CUET, Olympiad and custom, with sections, timing per section, negative marking and numeric or integer answers, plus question and option shuffling.
  • Available now: Online CBT on web and mobile: full screen with tab-switch warnings logged and no invasive proctoring, an NTA-style question palette, mark for review, one active session per student per test and screen capture blocked in the apps.
  • Available now: Resume after a power or internet cut with answers and remaining time preserved. Mass submissions at the deadline are queued and retried, so no answer is lost.
  • Available now: Offline OMR: generate OMR sheets as PDF per test, scan by phone camera or a document-feeder scanner, route low-confidence sheets to a review queue and detect wrong roll or test-code bubbles.
  • Available now: Answer-key publication and challenges; a bonus or changed key causes re-evaluation and re-rank. Ranks, percentiles and tie-breaks you set (section order, fewer wrong answers, accuracy).
  • Available now: Analysis: chapter-wise accuracy, time per question, attempt strategy, concept versus calculation errors and a comparison with the batch average and anonymised toppers.
  • Available now: A parent report as a PDF on WhatsApp after each test. Scores, ranks and chapter accuracy go automatically; AI-written notes go only after teacher approval.
  • Available now: Scholarship and admission tests with a public link, OTP login for guest leads and an automatic slab to offer.
  • Available now: Adaptive practice built from each student error log and question difficulty calibrated from response data.
  • Available now: Printed test booklets with sets A, B, C and D, and the opt-in public scholarship-test leaderboard that respects privacy.
  • Planned: Lab-network offline CBT using a local server.

What does the AI do here, and who approves it?

  • A question-paper generator builds a paper from a blueprint of chapters, marks, difficulty and competency percentages, with sets, an answer key and a marking scheme in LaTeX and in both languages. A coordinator approves it before use.
  • Question tagging by chapter, topic, Bloom level and difficulty, and duplicate detection, with spot checks.
  • Test analysis explains concept versus calculation errors and a time strategy for students and teachers; text sent to parents waits for teacher approval.
  • The student tutor is locked during a test.

How does it work?

How does Acavya run a coaching test series with CBT, OMR and ranks?

  1. Build the test

    Pick a pattern, choose questions or generate a paper from a blueprint, and set timing and marking.

  2. Conduct

    Students take the CBT, or sit the paper and the OMR sheets are scanned by phone.

  3. Check and rank

    Review low-confidence sheets, publish the key, handle challenges and re-rank.

  4. Report

    Analysis goes to students and teachers, and the parent report goes on WhatsApp.

Does it work in Hindi, offline and on WhatsApp?

  • Reports go to parents in Hindi or English and the student interface is bilingual.
  • CBT is built for patchy connections and power cuts; mass submission at the deadline is queued.
  • OMR scanning needs only a phone camera for small centres; a document-feeder scanner is optional.
  • Mathematics renders with LaTeX in questions and solutions.

What can it not do?

  • Our OMR accuracy target is 99 percent on an external phone-photo sample set; that external set is still pending, so treat it as a target.
  • No public rank lists by default, and batch formation is not merit-based, in line with the Ministry of Education coaching guidelines and the Supreme Court direction on state rules; check your state pack.
  • Tab-switch logging is a warning, not full proctoring.
  • The hints-first student tutor is part of the AI copilot and runs on free models inside your plan’s AI actions; it is not an extra product.
  • Results never promise ranks to students or parents.

What happens if the power goes during a CBT?

The answers are cached on the device, and when the student reconnects the test resumes with the remaining time.

Can I scan OMR sheets with a phone?

Yes. A phone camera or a document-feeder scanner works, and sheets the system is unsure about go to a review queue.

How does re-ranking work after a bonus question?

You change the key and publish it; Acavya re-evaluates every submission and recomputes ranks, percentiles and tie-breaks.

Does a free student tutor come with it?

Yes, in every paid plan. The hints-first tutor is part of the AI copilot and runs on free models inside your plan’s AI actions; with your own AI key it is unlimited.

Are ranks public?

No. Results are private by default so you do not publish rank lists unless you choose to.

One flat price, every feature: nothing here is an add-on. Find your plan.

30-day free trial for up to 30 students, no card needed. You set up your own institute, with an optional sample setup to explore. The Free plan is for tutors.

Start 30-day trialBook a 10-minute demo