TNPSC Group 4 Cut Off Marks: How the Cut-off Really Works (Caste-wise, Out of 300)

1. One rank list on total marks.

2. Onscreen certificate verification. TNPSC shortlists by rank and reservation: 1:3 for the general turn and 1:2 for reserved categories for VAO, Junior Assistant, Typist and Steno posts, and 1:6 for forest posts.

3. Physical certificate verification and counselling. VAO, Junior Assistant and Steno candidates are called at 1:3 (general) and 1:1.5 (reserved), Typists at 1:3 and 1:2. You pick a post in rank order when your turn comes.

So a cut-off is a result, not a rule: it is the mark of the last person reached in each slot. And one exam has a lot of slots.

Every post is filled through these communal turns, each with its own last rank:

Inside each turn there are further separate counts:

• Women: at least 30% of the vacancies in every turn. Women also compete for the other 70%.

• Destitute widows: 10% of the women's share.

• PSTM: candidates who studied in Tamil medium from Class 1 have their own preferential share.

• Persons with benchmark disability: 4% on posts identified as suitable.

• Ex-servicemen: their own reserved turn.

That is why a single "BC cut-off" or "MBC cut-off" on a coaching table is a simplification: it squeezes BC general, BC women, BC PSTM and the rest into one number, when each of them can end at a different rank.

Post-wise: VAO, Junior Assistant and Typist are separate races

The notification applies the reservation rule to each post separately. In 2026 that means 3,792 Junior Assistant posts, 1,572 Typist posts, 376 Steno-Typist posts and 291 VAO posts, each with its own roster. Typist posts need a government typewriting pass and Steno posts shorthand too, so their candidate pool is much smaller than for VAO or Junior Assistant, and the last rank called for them can look very different even within one community.

• Vacancies. More posts means TNPSC reaches deeper into the rank list, which pulls the cut-off down. 2024 ended with roughly 9,500 posts; 2025 with about 5,300.

• Posts added late. Additions often come after the exam. In 2025, 645 posts were added on 3 December, after the marks had been announced (Indian Express); in 2024, 41 were added in January 2025, after the October result. Every late addition lowers the effective cut-off.

• Candidates and difficulty. An easier paper or a bigger field pushes marks up at every rank, and the cut-off rises with them.

The paper has 200 questions for 300 marks, so every question carries 1.5 marks, in Part A and in Parts B and C alike.

• Correct answer: +1.5

• Wrong answer: 0

• Option [E] "answer not known": 0

• Question left blank: −0.5 each

• No left thumb impression on the OMR sheet: −0.5

Your score = 1.5 × correct answers − 0.5 × blank answers.

A worked example:

• Part A: 80 correct, so 80 × 1.5 = 120 marks (above 60, so Parts B and C are evaluated)

• Parts B and C: 62 correct, so 93 marks

• Total before deductions: 213

• 5 questions left blank: 5 × 0.5 = 2.5 off, so the final score is 210.5 of 300

Had those 5 bubbles been shaded E, the score would have been 213. Had they been guessed, it would have been at least 213.

The Part A minimum of 60 marks is 40 correct answers out of 100. The notification says blank-answer deductions come off your total, without spelling out how they interact with the Part A minimum, so keep a clear margin above 40 there and never leave a bubble empty.

Nobody can tell you the 2026 cut-off before the result: not us, and not a coaching centre. What you can control is your margin.

• The floor is 90. With more than 200 candidates per post, the people who reach certificate verification are typically well above it.

• A practical training target: in full-length timed mocks, aim to get about three-quarters right, around 150 of 200 correct (225 marks), with Part A your strongest part at 80 or more of 100. This is not a predicted cut-off. It is a buffer, because exam-day nerves and an unfamiliar paper can take a few marks off.

• Track three numbers per mock: your Part A score, your Parts B and C score, and blanks, which should always be zero.

• Lean on Part A. It is half the paper and the gate for the rest; our General Tamil guide breaks down where those marks come from.

Thervu's Group 4 tests are timed, in Tamil and English, and marked the moment you submit, with a score for each subject and the topics that cost you marks. Wrong answers come back in your revision queue until you get them right. For a week-by-week schedule to 10 January, use the 90-day study plan, and see the Group 4 syllabus for every unit.

Coaching sites publish "expected" or "previous year" caste-wise cut-offs. Some may be built from marks and ranks that candidates report to them, but none we checked links to a TNPSC document, and the figures differ from site to site. Treat them as rough guesses. If a page calls a figure official, look for the tnpsc.gov.in document behind it.

The real evidence is what TNPSC itself publishes on tnpsc.gov.in: the written result, then the lists of candidates called for certificate verification and counselling. Those lists, not coaching tables, are the closest public evidence of how far down the rank list TNPSC went. We will add what TNPSC publishes for the 2026 exam here as it comes out. For the exam itself, the TNPSC Group 4 exam guide has the pattern and eligibility.