[[48,18,4]] · QEC Challenge

← back to the board

[[48,18,4]] d ≤

kd²/n

6.0

X/Z

1.25

0.0089

5.099

layers

swaps

Share this result

Distance

X/Z asymmetry 1.25 · d_X ≤ 4, d_Z ≤ 5 · w_X = 17, w_Z = 18 (max(d_X,d_Z)/min(d_X,d_Z); each side carries its own earned tier: = certified exact, ≤ witness upper bound)

d_X 4 · witness weight 4 (claimed upper_bound)

witness operator (support, 4 qubits)

[10, 16, 24, 29]

d_Z 5 · witness weight 5 (claimed upper_bound)

witness operator (support, 5 qubits)

[5, 10, 23, 34, 45]

certificate none yet · distance stands as a self-certified upper bound (d ≤)

Diagnostics

computed by the verifier from the parity checks, the layout, and the stored witnesses; shown as evidence, not used for ranking

girth H_X 4 · H_Z 4 (shortest cycle of each side’s Tanner graph; longer is friendlier to belief propagation)

check weights H_X 9–17 (mean 11.438) · H_Z 16–18 (mean 17.143)

qubit degrees H_X 1–8 (mean 3.812) · H_Z 1–9 (mean 5.0)

trapping sets H_X (1,1)×7 (2,1)×5 (3,1)×20 (smallest syndrome weight at each size, connected sets of up to 3 qubits)

full (size, syndrome weight): count census for H_X

(1,1): 7 (1,2): 7 (1,3): 12 (1,4): 5 (1,5): 4 (1,6): 6 (1,7): 6 (1,8): 1 (2,1): 5 (2,2): 29 (2,3): 39 (2,4): 89 (2,5): 102 (2,6): 136 (2,7): 109 (2,8): 79 (2,9): 38 (2,10): 10 (2,11): 2 (2,12): 1 (2,13): 1 (3,1): 20 (3,2): 100 (3,3): 316 (3,4): 736 (3,5): 1401 (3,6): 1775 (3,7): 1934 (3,8): 1626 (3,9): 1011 (3,10): 476 (3,11): 211 (3,12): 71 (3,13): 16 (3,14): 2

trapping sets H_Z (1,1)×5 (2,1)×3 (3,1)×21 (smallest syndrome weight at each size, connected sets of up to 3 qubits)

full (size, syndrome weight): count census for H_Z

(1,1): 5 (1,2): 3 (1,3): 5 (1,4): 4 (1,5): 8 (1,6): 11 (1,7): 7 (1,8): 2 (1,9): 3 (2,1): 3 (2,2): 16 (2,3): 26 (2,4): 88 (2,5): 124 (2,6): 157 (2,7): 183 (2,8): 148 (2,9): 80 (2,10): 29 (2,11): 8 (2,12): 2 (2,13): 1 (3,1): 21 (3,2): 97 (3,3): 347 (3,4): 1002 (3,5): 1768 (3,6): 2715 (3,7): 2917 (3,8): 2364 (3,9): 1536 (3,10): 718 (3,11): 281 (3,12): 63 (3,13): 4

witness diameter X 4.4721 · Z 3.6056 (Euclidean support diameter of the stored distance witnesses in the layout; an upper bound on the exhibited logicals’ spread, not a minimum over all logicals)

Verified 2D layout

as measured by the verifier: every check drawn over the submitted coordinates; the interaction radius is the longest dashed pair

layout contributed by @dorakingx · Simulated annealing over site assignments on an integer grid with the repo's own research/local2d/fold_layout.search_layout (max-diameter objective, several bounding boxes and seeds); single layer at radius 4 tried first, then two layers at 7. Radius and per-site occupancy re-measured with the verifier's rule. · 2026-10-02

X checkZ checkqubit site (24)2 qubits stacked (2 layers)dashed: the pair setting the interaction radiushover a check to isolate its qubits; click to pin — repeated clicks cycle through overlapping checks; click empty space to release

routing cost 49 nearest-neighbor SWAPs per round in total, at most 5 for one check (heuristic: MST lower bound on the layout, with one lattice step = the minimum qubit spacing 1; not a rank)

Construction & provenance

authors @vprusso

provenance submitted through the challenge

novelty known parameter set; see provenance notes

construction Check deletion from the board entry [[48,16,5]] (codes/48-16-5.json): rows [6, 8] of H_Z removed. Dropping two independent stabiliser generators frees two logical qubits and cannot raise the distance.

model Claude Opus 5 (claimed, not verified)

date 2026-09-21

notes Found by sweeping every single-row deletion of every small board entry and keeping those that land undominated. The parent's remaining checks are unchanged.

family check-deletion (a tag, not a ranking)

locality 2D-local bilayer (computed from the layout)

weight class weight > 8 (computed)

How this code was found

the research note submitted with this code · raw markdown · all notes

[[48,18,4]] by two-row check deletion from 48-16-5

Hypothesis

Every stabiliser generator costs one logical qubit, so dropping two independent generators returns k + 2 on the same n physical qubits at a distance that can only fall.

Whether the trade is worth taking

Usually not. Across 5,033 pair deletions of board entries, 12.5% kept a distance of 3 or more, against 40% for single-row deletion on the same parents. Dropping a second generator costs far more distance than the first. This code is one of the few where it lands undominated anyway, in a k + 2 region of its cell that nothing on the board covers.

Construction

Parent: codes/48-16-5.json. Rows [6, 8] of H_Z are removed; every other check is untouched. The pair is checked for independence (the rank must fall by exactly two) and the result for Tanner connectivity before anything else runs.

Distance

Two-sided random information-set search, then the heavier side on its own in the quotient ker(H_opposite) / rowspace(H_own), since a two-sided search returns the minimum over both sides and never surfaces the heavier one. Deleting rows leaves H_X and H_Z with different shapes, so every witness is classified by the conditions it actually satisfies rather than by the search's argument order.

Boundary

Distance is an upper bound from witness search, not a proof. The parent's own distance claim is inherited context. Literature novelty is unverified.

Parity checks

X-checks 16 (max weight 17) · Z-checks 14 (max weight 18)

H_X (16 checks, sparse supports)

[0, 2, 6, 10, 13, 23, 28, 46, 47] [1, 5, 9, 12, 22, 27, 45, 46, 47] [3, 4, 6, 7, 14, 18, 21, 26, 29, 41, 42, 43, 44, 46, 47] [3, 8, 10, 20, 25, 43, 44, 45, 47] [0, 3, 6, 7, 8, 9, 12, 13, 21, 30, 37, 38, 40, 43] [0, 1, 3, 7, 11, 14, 24, 29, 47] [0, 1, 4, 5, 8, 13, 17, 19, 23, 37, 38, 39, 41, 42] [1, 7, 13, 16, 19, 22, 37, 38, 39, 40, 41, 43, 44, 47] [1, 3, 4, 6, 13, 15, 19, 21, 37, 38, 40, 42, 44, 46] [2, 5, 14, 20, 38, 39, 40, 41, 45] [0, 3, 12, 18, 36, 37, 38, 40, 43] [2, 11, 17, 35, 36, 37, 39, 42, 47] [1, 10, 16, 34, 35, 36, 38, 41, 46] [0, 1, 4, 9, 13, 15, 19, 33, 34, 35, 38, 39, 44, 45] [0, 3, 6, 7, 11, 18, 29, 32, 33, 34, 36, 40, 41, 42, 43, 44, 45] [8, 13, 31, 32, 33, 35, 39, 43, 46]

H_Z (14 checks, sparse supports)

[0, 4, 5, 7, 8, 11, 12, 13, 14, 17, 24, 25, 29, 30, 32, 33] [3, 4, 6, 7, 8, 10, 11, 12, 13, 16, 23, 24, 28, 29, 31, 32, 47] [2, 3, 5, 6, 7, 9, 10, 11, 12, 15, 22, 23, 24, 27, 28, 30, 31, 46] [1, 2, 4, 5, 6, 9, 10, 11, 14, 21, 22, 23, 26, 27, 29, 30, 45] [0, 1, 3, 4, 5, 8, 9, 10, 13, 20, 21, 22, 25, 26, 28, 29, 44] [0, 2, 3, 4, 7, 8, 9, 12, 19, 20, 21, 25, 27, 28, 43, 47] [0, 1, 2, 5, 6, 7, 8, 10, 17, 18, 19, 23, 24, 25, 26, 41, 45, 46] [0, 3, 4, 5, 6, 15, 16, 17, 21, 22, 24, 40, 43, 44, 46, 47] [2, 3, 4, 5, 8, 14, 15, 16, 20, 21, 23, 24, 39, 42, 43, 45, 46, 47] [1, 2, 3, 4, 7, 13, 14, 15, 19, 20, 22, 23, 38, 41, 42, 44, 45, 46] [0, 1, 2, 3, 6, 12, 13, 14, 18, 19, 21, 22, 37, 41, 43, 44, 45] [0, 1, 2, 5, 11, 12, 13, 17, 18, 20, 21, 36, 40, 42, 43, 44, 47] [0, 1, 4, 10, 11, 12, 16, 17, 19, 20, 35, 39, 40, 41, 42, 43, 46, 47] [0, 3, 9, 10, 11, 15, 16, 18, 19, 34, 38, 39, 41, 42, 45, 46, 47]

Code ID 48-18-4 · download JSON · raw on GitHub