Millennium Problems / 01

Riemann Hypothesis

First 4,000 zeros, re-verified: two independent implementations, every zero on the critical line, max real-part deviation 0.0.

Program
The Millennium Lab
Rounds
1, 2, 4, 5
Zeros
4,000
Receipts
66, head bdebd1e7
Date
Nov 20-21, 2026
Status
NOT SOLVED
Method
PARI/GP 2.19.1 + mpmath
Top of range
t = 4506.31
NOT SOLVED. Nothing on this page solves the problem, proves it, or disproves it. This is research: measurements, rejections, and partial formalizations, each with a receipt. The problem belongs to the mathematicians who stated it and the researchers who have worked it for decades. This program adds a small computational exploration on top of their work and claims no credit for theirs.

01The problem, in plain words

The Riemann zeta function zeta(s) has zeros at the negative even integers (the trivial zeros) and infinitely many others, the non-trivial zeros, which all lie in the strip 0 < Re(s) < 1. Riemann conjectured in 1859 that every non-trivial zero sits exactly on one vertical line, where the real part equals one half. If true, the primes are distributed as regularly as the analytic structure of zeta permits. If false, a single zero off the line breaks the pattern. The conjecture is a statement about all infinitely many zeros; no finite check can prove it.

02What the lab actually ran

Two independent implementations, run against each other on every zero. First, PARI/GP 2.19.1 lfunzeros locates every zero with imaginary part in the scan interval (C implementation, Riemann-Siegel, argument principle). Then, for each located t, an independent mpmath Newton step from 0.5 + i t, using zeta and zeta prime at 30 decimal digits. The Newton step is free to move off the line; the reported real part is where it converges. The two implementations agree: the first zero from PARI matches the mpmath bisection value to 22 digits. A count check against the Riemann-von Mangoldt formula N(T) at the top of the interval confirms no zero was missed.

RoundZerosInterval Last tMax |zeta| at line von Mangoldt expectedS(T)
1 (Nov 20)1,000[0, 1420] 1419.42255.6087e-26999.420.582
2 (Nov 20)2,000[0, 2560] 2515.28656.5408e-261999.380.620
4 (Nov 20)3,000[0, 3550] 3533.32826.5096e-212999.370.626
5 (Nov 21)4,000[0, 4650] 4506.31157.4547e-213999.530.469

In every round the largest deviation of a zero's real part from 1/2 was 0.0 at 30-digit precision: every Newton step returned Re = 1/2 exactly at this precision, so the deviation is below 1e-30. Earlier rounds reproduce exactly: the first 1,000 t values match round 1, the first 2,000 match round 2, the first 3,000 match round 4. Round 5 added 1,000 new zeros beyond round 4: 984 from the new interval [3545, 4650] plus 16 leftover zeros from the round 4 upper scan. The new scan ran twice with identical hashes; the second run took 414.0 seconds and the mpmath verification pass took 560.91 seconds. Scan times for the upper intervals: round 2, 907.3 s; round 4, 1699.8 s.

11252250337545001000Round 1t to 1419.422000Round 2t to 2515.293000Round 4t to 3533.334000Round 5t to 4506.31zeros verified on the critical line
Figure 1. Zeros verified on the critical line per round: 1,000, then 2,000, then 3,000, then 4,000. Each round reuses the earlier zeros and adds a new interval on top. Computed from riemann_zeros.json, riemann_zeros_r2.json, riemann_zeros_r4.json, riemann_zeros_r5.json.
The critical line Re(s) = 1/2, with the first 10 zeros marked by their imaginary part t.Re(s) = 1/2t = 0t = 10t = 20t = 30t = 40t = 50n=114.13n=221.02n=325.01n=430.42n=532.94n=637.59n=740.92n=843.33n=948.01n=1049.77
Figure 2. The first 10 non-trivial zeros, plotted on the critical line Re(s) = 1/2 by their imaginary part t. Every one of the 4,000 zeros checked sits on this line by the same test; the figure shows the first 10 because the line is the point. Values from riemann_zeros_r5.json.

03The first 10 zeros

nt (imaginary part) Re(s) from NewtonRe deviation |zeta(1/2 + i t)|
114.134725141734693790457250.50.01.5733e-24
221.022039638771554992628480.50.04.6167e-25
325.010857580145688763213790.50.01.3615e-24
430.42487612585951321031190.50.03.22e-24
532.935061587739189690662370.50.01.4318e-24
637.586178158825671257217760.50.06.7404e-24
740.918719012147495187398130.50.04.5991e-24
843.327073280914999519496120.50.03.9703e-24
948.005150881167159727942470.50.04.3112e-24
1049.773832477672302181916780.50.06.6386e-24

All 4,000 zeros are listed in the run files, not only the first 10 shown here. Results file SHA-256 (round 5): 720c758471012ad6107678bffe23bd80b9d73bbb9ef57d7cacd9a85a89a3c73e.

04What we found

05Limits and blind spots

06Receipts

Receipt chain: 66 receipts, chain valid, head bdebd1e7d93dd108bffe0ba01d499e603db81a8c17027bb0aad36a70f73dacdd (verification receipt rcpt_0066). Riemann receipts:

ReceiptWhat it covers
rcpt_0001Round 1: first 1,000 zeros on the critical line
rcpt_0042Round 2 calibration: first 2,000 zeros
rcpt_0052Round 4 calibration: first 3,000 zeros
rcpt_0058Round 5 calibration: first 4,000 zeros

Data files: data/riemann_zeros.json, riemann_zeros_r2.json, riemann_zeros_r4.json, riemann_zeros_r5.json; verification logs riemann_verify.log, riemann_verify_r4.log.

07Next attack

More zeros (10,000) on the same two-implementation method, or the proportion direction: an independent check of the claimed 67 percent result once the manuscript is released. The lab's method checks zeros one by one; the proportion result is a different kind of theorem and would need its own attack.

08What we found, in plain language

We checked the first 4,000 interesting zeros of the zeta function with two different programs, and every one of them sits exactly on the line Riemann predicted in 1859. The count of zeros also matches the formula that predicts how many there should be, so we are fairly confident none was missed. That is a careful re-check of known ground, not a step toward a proof.

09What we did not find

No zero off the line, no proof, and no new record. We did not check any zero past t = 4506.31, and we did not touch the question of what proportion of all zeros lie on the line.

The Millennium Lab. Research pages, November 2026. All numbers trace to the lab data files; every receipt is listed on the page. Nothing solved.