Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Log inSign up
Challenges
Datasets
Workspaces
Papers
Discussions
Leaderboard
Blog
Job Board
Q3AS

© 2026 Aqora Quantum S.A.S.

TermsPrivacyLegal Notice

Distance-scaling fault tolerance of the three-sheet multiplexed FCC sheet code at L = 4, 6, 8

Raghu Kulkarni

Published Jun 11, 2026• 0 citations
Publisher

Abstract

Abstract We provide a circuit-level numerical validation of the three-sheet time-multiplexed FCC sheet code, a quantum low-density parity-check (qLDPC) construction with parameters [[3L 3 , 6L, L]] derived from the three triad sheets of the face-centered cubic lattice. At L = 4 the code packs 24 logical qubits into 192 data qubits at distance 4, an encoding rate of 12.5%, higher than any other published code at this distance. Three independent sheet-codes share one chip; ancillas are shared, stabilizers are extracted in three sequential rounds, and the inactive sheets accumulate depolarizing noise during each round at rate 2p. We simulate the full circuit-level memory experiment with explicit CNOT cascades, depolarizing noise on every gate, and minimum-weight perfect matching decoding via PyMatching on Stim's detector error model. Threshold under the multiplex idle penalty falls between 0.4% and 0.5%, in agreement with the original construction paper. Distance scaling is sharp: the per-logical error rate suppresses by Λ(4→6) = 11.2 between consecutive even L at p = 0.001, and Λ(6→8) = 7.8. At L = 8 the [[1536, 48, 8]] code achieves a per-logical error rate of 5.9 × 10⁻⁵, more than an order of magnitude below the physical rate, on a 48-logical-qubit memory. These suppression factors, measured in simulation deeply sub-threshold, are over five times those recently reported on noisy hardware on a single-logical surface-code memory; an apples-to-apples comparison at matched operating point requires either future hardware experiments on this code or a surface-code simulation at the same deeply sub-threshold operating point and is not made here. The natural hardware target is the L = 6 instance: [[648, 36, 6]] encodes 36 logical qubits at distance 6 in 864 physical qubits, fitting Atom Computing's 1180-qubit Phoenix processor and exceeding every prior multi-logical hardware demonstration in distance and logical-qubit count. The L = 4 instance overlaps with Atom Computing and Microsoft's 24-logical [[4, 2, 2]] demonstration, providing a calibration point at quadrupled distance (d = 4 vs d = 2) and full error correction in place of error detection.

DOI
10.21203/rs.3.rs-9737123/v1

Related papers

Ancilla-Shared Time Multiplexing of Three FCC Sheet Codes: A One-Third Qubit Saving at a 1.3× Threshold Cost

Raghu Kulkarni

The three triad sheets of the face-centered cubic (FCC) lattice, at even lattice size L, are edge-disjoint but share their vertex and octahedral-void ancilla positions. Running the three sheet codes in a three-round time-multiplexed cycle on one chip needs 4L3 physical qubits against 6L3 for three separate blocks, a saving of exactly one third at every L, for instance, 864 qubits instead of 1296 at L=6. The schedule is implemented explicitly, with all three sheets on one shared ancilla set; comparing it against the effective single-sheet model used for the parameter sweeps shows agreement to within 5% at L=4 and 2.2% at L=6. The cost of the saving is that each sheet idles for two of every three sub-rounds, and the finite-size crossing estimate of the circuit-level threshold falls from 1.13±0.08% to 0.86±0.06%: a factor of 1.3, that is a 24% relative reduction, or 0.27 percentage points. Sweeping the idle-to-gate noise ratio r over 0,1,2,4 gives estimates decreasing monotonically from 1.13% to 0.71%. The dual X-basis memory gives 0.78±0.05%, overlapping the Z value. All values are finite-size crossings at L=4,6,8 from the full FCC circuit, not asymptotic thresholds, and the quoted spreads are dominated by finite-size drift rather than statistical error. Each sheet decomposes exactly into L independent rotated 2D toric codes (Proposition 1), so the memory is the toric code’s and no encoding-rate advantage is claimed; at d=4, the rate equals the rotated toric code’s. In the simulated noise model, where no channel couples two sheets, the sheets fail independently. Sub-threshold suppression survives the sharing at Λ(4→6)=12.3±1.1 and Λ(6→8)=5.6±1.2 at p=10−3, on a memory carrying 24 to 48 logical qubits. Block-level failure probabilities are reported alongside, since they grow with the logical count: at L=6 and p=10−3, the 36-logical chip loses at least one logical qubit in about one experiment in sixty. The L=6 instance, [[648,36,6]], needs 864 physical qubits, within the nominal atom count of current neutral-atom processors; no implementability claim is made beyond that count.

Quantum Reports•Sep 21, 2026•0 citations

Three Sheets, One Architecture: Inter-Sheet Logical Gates for K=4 Quantum Error Correction

Raghu Kulkarni

Abstract We present a CSS quantum error correcting code on the Face-Centered Cubic (FCC) lattice that combines surface-code-compatible K=4 connectivity with native inter-sheet logical gates between co-located CSS code blocks — a combination not previously achieved among well-studied codes. The construction has two parts. First, restricting the [[3L³, 2L³+2, 3]] FCC lattice code to a single triad sheet yields the sheet code with parameters [[L³, 2L, L]] at even L (or [[L³, L, L]] as a planar variant), uniform weight-4 stabilizers, and K=4 per-qubit connectivity. Three triad sheets share the FCC lattice geometry and can be deployed in one of three ways — monolithic 2D for small L, a three-layer stacked architecture for scalable deployment, or native 3D hardware — in each case encoding 6L logical qubits at distance L. Second, we introduce a fault-tolerant lattice surgery protocol that uses local FCC triangle measurements to implement joint Pauli measurements between logical qubits in different sheets. Each surgery operation adds O(L) two-qubit gates to the O(L³) per-round syndrome extraction cost. Under circuit-level depolarizing noise with MWPM decoding, the static memory threshold is p_th ≈ 0.63% and the surgery threshold is p_th ≈ 0.5%, both above the operating range of current quantum processors. Compared to surface code lattice surgery (also K=4, lattice surgery confined within a single patch), the sheet code provides 2L logicals per sheet at the same distance and connectivity; compared to bivariate bicycle codes (which target inter-block gates between physically separated modules at K=6 with long-range couplers), the sheet code achieves inter-sheet gates at K=4 with only short-range couplings between co-located sheets.

•Jun 19, 2026•0 citations

Tunable Distance Quantum Error Correction at 29 Percent Encoding Rate of the Face Centered Cubic Lattice

Raghu Kulkarni

Abstract The [[3𝐿 3 , 2𝐿 3 +2, 3]] CSS code on the Face-Centered Cubic (FCC) lattice achieves 67.7% encoding rate with weight-12 stabilizers, but its distance is capped at 𝑑 = 3 by weight-3 logical operators sitting at tetrahedral interstitial voids. These logicals carry a geometric fingerprint—one error per triad sheet, all three converging at a known void—that makes them detectable by a flag ancilla measuring the three-body parity 𝑍 𝑒𝑎 𝑍 𝑒𝑏 𝑍 𝑒𝑐 . The weight-3 logical flips this parity from +1 to −1; the stabilizer syndrome stays silent, but the flag sees it. Running 𝑅 independent flag rounds per syndrome cycle pushes the logical failure rate to 𝑝 3+𝑅 , so the effective distance is simply 𝑑eff = 3+𝑅—adjustable at runtime, with no changes to the code or its logical count. At 𝑅 = 4 the FCC+flags code achieves the same 𝑝 7 distance scaling as a distance-7 surface code, with a prefactor 𝑁void = 256 versus ∼ 10 for the surface code—the cost of encoding 130 logical qubits at 29.0% rate versus 1 logical qubit at 1.0%. An adaptive scheduling protocol cuts the flag overhead to under 0.1% of measurement cycles at realistic noise levels.

•May 15, 2026•0 citations

A 67%-Rate CSS Code on the FCC Lattice: [[192,130,3]] from Weight-12 Stabilizers

Raghu Kulkarni

We construct a three-dimensional Calderbank-Shor-Steane (CSS) stabilizer code on the Face-Centered Cubic (FCC) lattice. Physical qubits reside on the edges of the lattice (coordination $K=12$); X-stabilizers act on octahedral voids and Z-stabilizers on vertices, both with uniform weight 12. Computational verification confirms CSS validity ($H_{X}H_{Z}^{T}=0$ over GF(2)) and reveals $k=2L^{3}+2$ logical qubits: $k=130$ at $L=4$ and $k=434$ at $L=6$, yielding encoding rates of 67.7% and 67.0% respectively. The minimum distance $d=3$ is proven exactly by exhaustive elimination of all weight-$\le 2$ candidates combined with constructive weight-3 non-stabilizer codewords. The code parameters are [[192, 130, 3]] at $L=4$ and [[648, 434, 3]] at $L=6$. This rate is 24x higher than the cubic 3D toric code (2.8% at $d=4$), though at a lower distance ($d=3$ vs. $d=4$); the comparison is across different distances. The high rate originates in a structural surplus: the FCC lattice has $3L^{3}$ edges but only $L^{3}-2$ independent stabilizer constraints, leaving $k=2L^{3}+2$ logical degrees of freedom. We provide a minimum-weight perfect matching (MWPM) decoder adapted to the FCC geometry, demonstrate a 10x coding gain at $p=0.001$ (and 63x at $p=0.0005$), and discuss implications for fault-tolerant quantum computing on neutral-atom and photonic platforms.

quant-pharXiv:2603.20294•Mar 20, 2026•5 citations

A Hybrid Atom-Array Architecture for Universal Quantum Computation: The FCC Sheet Code with 3D Color-Code Magic-State Factory

Raghu Kulkarni

A fault-tolerant quantum computer needs a memory code and a source of magic states. On a neutral-atom array the two can use different codes. This paper connects a [[15,1,3]] color-code factory, which has a transversal T gate, to the FCC sheet code, a [[3L³, 6L, L]] CSS memory at K = 4 connectivity. The connection is a transfer: a logical state is moved from a color-code block into one sheet-code logical by one-bit teleportation. We build the transfer circuit with a flagged joint Z̄ₐZ̄ᵦ measurement and simulate it at L = 4 under circuit-level depolarizing noise. The noise-free truth table holds for all four basis states. The circuit distance is 4 in the Z sector and 3 in the X sector. At p = 10⁻³ the transfer fails at 0.17% (Z) and 1.7% (X); the sheet memory alone fails at 0.07% and below 0.27% over the same rounds. When all noise on the color-code block is turned off, the X-sector excess disappears. The cost of the transfer is therefore the distance-3 factory block failing while it waits, not the interface. Two further results support the architecture. The lattice rotation R: (x, y, z) ↦ (y, z, x) maps each of the three sheets onto the next. We verify at L = 4, 6, 8 that R is a CSS isomorphism and that it permutes the 2L logicals of a sheet. The sheets are copies of one code with a pairing fixed by the lattice, and the standard transversal CNOT between copies of a CSS code applies to any two sheets. One CNOT layer on R-paired qubits gives 2L logical CNOTs. Under noise, decoding the two sheets separately fails on the target sheet, with an apparent threshold near 0.3%, because the CNOT copies the control sheet's residual errors onto the target. A feed-forward decoder that propagates the decoded control errors through R restores suppression with L, with a crossing near 0.8% against the memory's 1.1%; with exact knowledge of the control errors the memory value is recovered. We also count the atoms used by the cross-sheet triangle-ribbon merge of the companion surgery paper: 3L/2 against 2L for a surface-code routing channel. We state what is not new. Each sheet is L rotated toric codes, so the memory is the toric code's. A transversal CNOT between copies of a CSS code is available to any stacked code, including surface-code patches on atom arrays. The sheet code adds a lattice-fixed pairing and the ribbon merge, not a throughput advantage; the pairing is not spatially local, since the distance between R-paired qubits grows as about 0.47L.

•Jun 17, 2026•0 citations

High-rate qLDPC processors

Aditya Bhardwaj, Muzhou Ma, Nadine Meister +6 more

Despite significant progress on quantum low-density parity-check (qLDPC) codes, building qLDPC processors that are high-rate, high-throughput, hardware-friendly, and fast-to-decode remains a challenge. We introduce mitten codes, a family of qLDPC processor codes of encoding rate $20\%$ and check weight $9$, based on non-abelian groups. Their non-abelian structure evades distance bounds constraining abelian counterparts, allowing mitten codes to reach distance $18$ and beyond with just a few hundred data qubits. The logical operators of a mitten code are related by the group action, yielding a modular, low-overhead logical toolkit: full Clifford operations follow from bridging two reusable seed surgery gadgets or from a single fixed extractor. Furthermore, qLDPC processors based on mitten codes support high-rate surgery that executes many logical measurements in parallel, and parallel magic-state injection into all logical qubits at once. Under circuit-level noise, with our fast decoder, the $[\![300,60,14]\!]$ mitten code achieves, without extrapolation, a block logical error rate of ${\sim}10^{-11}$ per round at $0.1\%$ physical error rate (PER), while the $[\![ 975,195,\leq 24 ]\!]$ code reaches ${\sim}10^{-8}$ at $0.4\%$ PER. Decoding $15$ billion surgery experiments on the $[\![540,108,18]\!]$ code at $0.1\%$ PER, we observe only two logical failures, demonstrating a qLDPC processor capable of running ${\sim}10^{10}$ logical operations. Our decoder is compatible with sub-millisecond average latency per logical cycle, sufficient for real-time decoding on neutral atom hardware. Discovered by an end-to-end design pipeline built on sQetch, a distance estimator orders of magnitude faster than existing tools, and mapping efficiently onto near-term neutral atom and superconducting hardware, mitten codes open a practical path toward fault-tolerant quantum computation.

quant-pharXiv:2607.28795•Aug 3, 2026•45 citations
Abstract We provide a circuit-level numerical validation of the three-sheet time-multiplexed FCC sheet code, a quantum low-density parity-check (qLDPC) construction with parameters [[3L 3 , 6L, L]] derived from the three triad sheets of the face-centered cubic lattice. At L = 4 the code packs 24 logical qubits into 192 data qubits at distance 4, an encoding rate of 12.5%, higher than any other published code at this distance. Three independent sheet-codes share one chip; ancillas are shared, stabilizers are extracted in three sequential rounds, and the inactive sheets accumulate depolarizing noise during each round at rate 2p. We simulate the full circuit-level memory experiment with explicit CNOT cascades, depolarizing noise on every gate, and minimum-weight perfect matching decoding via PyMatching on Stim's detector error model. Threshold under the multiplex idle penalty falls between 0.4% and 0.5%, in agreement with the original construction paper. Distance scaling is sharp: the per-logical error rate suppresses by Λ(4→6) = 11.2 between consecutive even L at p = 0.001, and Λ(6→8) = 7.8. At L = 8 the [[1536, 48, 8]] code achieves a per-logical error rate of 5.9 × 10⁻⁵, more than an order of magnitude below the physical rate, on a 48-logical-qubit memory. These suppression factors, measured in simulation deeply sub-threshold, are over five times those recently reported on noisy hardware on a single-logical surface-code memory; an apples-to-apples comparison at matched operating point requires either future hardware experiments on this code or a surface-code simulation at the same deeply sub-threshold operating point and is not made here. The natural hardware target is the L = 6 instance: [[648, 36, 6]] encodes 36 logical qubits at distance 6 in 864 physical qubits, fitting Atom Computing's 1180-qubit Phoenix processor and exceeding every prior multi-logical hardware demonstration in distance and logical-qubit count. The L = 4 instance overlaps with Atom Computing and Microsoft's 24-logical [[4, 2, 2]] demonstration, providing a calibration point at quadrupled distance (d = 4 vs d = 2) and full error correction in place of error detection.
The three triad sheets of the face-centered cubic (FCC) lattice, at even lattice size L, are edge-disjoint but share their vertex and octahedral-void ancilla positions. Running the three sheet codes in a three-round time-multiplexed cycle on one chip needs 4L3 physical qubits against 6L3 for three separate blocks, a saving of exactly one third at every L, for instance, 864 qubits instead of 1296 at L=6. The schedule is implemented explicitly, with all three sheets on one shared ancilla set; comparing it against the effective single-sheet model used for the parameter sweeps shows agreement to within 5% at L=4 and 2.2% at L=6. The cost of the saving is that each sheet idles for two of every three sub-rounds, and the finite-size crossing estimate of the circuit-level threshold falls from 1.13±0.08% to 0.86±0.06%: a factor of 1.3, that is a 24% relative reduction, or 0.27 percentage points. Sweeping the idle-to-gate noise ratio r over 0,1,2,4 gives estimates decreasing monotonically from 1.13% to 0.71%. The dual X-basis memory gives 0.78±0.05%, overlapping the Z value. All values are finite-size crossings at L=4,6,8 from the full FCC circuit, not asymptotic thresholds, and the quoted spreads are dominated by finite-size drift rather than statistical error. Each sheet decomposes exactly into L independent rotated 2D toric codes (Proposition 1), so the memory is the toric code’s and no encoding-rate advantage is claimed; at d=4, the rate equals the rotated toric code’s. In the simulated noise model, where no channel couples two sheets, the sheets fail independently. Sub-threshold suppression survives the sharing at Λ(4→6)=12.3±1.1 and Λ(6→8)=5.6±1.2 at p=10−3, on a memory carrying 24 to 48 logical qubits. Block-level failure probabilities are reported alongside, since they grow with the logical count: at L=6 and p=10−3, the 36-logical chip loses at least one logical qubit in about one experiment in sixty. The L=6 instance, [[648,36,6]], needs 864 physical qubits, within the nominal atom count of current neutral-atom processors; no implementability claim is made beyond that count.
Abstract We present a CSS quantum error correcting code on the Face-Centered Cubic (FCC) lattice that combines surface-code-compatible K=4 connectivity with native inter-sheet logical gates between co-located CSS code blocks — a combination not previously achieved among well-studied codes. The construction has two parts. First, restricting the [[3L³, 2L³+2, 3]] FCC lattice code to a single triad sheet yields the sheet code with parameters [[L³, 2L, L]] at even L (or [[L³, L, L]] as a planar variant), uniform weight-4 stabilizers, and K=4 per-qubit connectivity. Three triad sheets share the FCC lattice geometry and can be deployed in one of three ways — monolithic 2D for small L, a three-layer stacked architecture for scalable deployment, or native 3D hardware — in each case encoding 6L logical qubits at distance L. Second, we introduce a fault-tolerant lattice surgery protocol that uses local FCC triangle measurements to implement joint Pauli measurements between logical qubits in different sheets. Each surgery operation adds O(L) two-qubit gates to the O(L³) per-round syndrome extraction cost. Under circuit-level depolarizing noise with MWPM decoding, the static memory threshold is p_th ≈ 0.63% and the surgery threshold is p_th ≈ 0.5%, both above the operating range of current quantum processors. Compared to surface code lattice surgery (also K=4, lattice surgery confined within a single patch), the sheet code provides 2L logicals per sheet at the same distance and connectivity; compared to bivariate bicycle codes (which target inter-block gates between physically separated modules at K=6 with long-range couplers), the sheet code achieves inter-sheet gates at K=4 with only short-range couplings between co-located sheets.
Abstract The [[3𝐿 3 , 2𝐿 3 +2, 3]] CSS code on the Face-Centered Cubic (FCC) lattice achieves 67.7% encoding rate with weight-12 stabilizers, but its distance is capped at 𝑑 = 3 by weight-3 logical operators sitting at tetrahedral interstitial voids. These logicals carry a geometric fingerprint—one error per triad sheet, all three converging at a known void—that makes them detectable by a flag ancilla measuring the three-body parity 𝑍 𝑒𝑎 𝑍 𝑒𝑏 𝑍 𝑒𝑐 . The weight-3 logical flips this parity from +1 to −1; the stabilizer syndrome stays silent, but the flag sees it. Running 𝑅 independent flag rounds per syndrome cycle pushes the logical failure rate to 𝑝 3+𝑅 , so the effective distance is simply 𝑑eff = 3+𝑅—adjustable at runtime, with no changes to the code or its logical count. At 𝑅 = 4 the FCC+flags code achieves the same 𝑝 7 distance scaling as a distance-7 surface code, with a prefactor 𝑁void = 256 versus ∼ 10 for the surface code—the cost of encoding 130 logical qubits at 29.0% rate versus 1 logical qubit at 1.0%. An adaptive scheduling protocol cuts the flag overhead to under 0.1% of measurement cycles at realistic noise levels.
We construct a three-dimensional Calderbank-Shor-Steane (CSS) stabilizer code on the Face-Centered Cubic (FCC) lattice. Physical qubits reside on the edges of the lattice (coordination K=12K=12K=12); X-stabilizers act on octahedral voids and Z-stabilizers on vertices, both with uniform weight 12. Computational verification confirms CSS validity (HXHZT=0H_{X}H_{Z}^{T}=0HX​HZT​=0 over GF(2)) and reveals k=2L3+2k=2L^{3}+2k=2L3+2 logical qubits: k=130k=130k=130 at L=4L=4L=4 and k=434k=434k=434 at L=6L=6L=6, yielding encoding rates of 67.7% and 67.0% respectively. The minimum distance d=3d=3d=3 is proven exactly by exhaustive elimination of all weight-≤2\le 2≤2 candidates combined with constructive weight-3 non-stabilizer codewords. The code parameters are [[192, 130, 3]] at L=4L=4L=4 and [[648, 434, 3]] at L=6L=6L=6. This rate is 24x higher than the cubic 3D toric code (2.8% at d=4d=4d=4), though at a lower distance (d=3d=3d=3 vs. d=4d=4d=4); the comparison is across different distances. The high rate originates in a structural surplus: the FCC lattice has 3L33L^{3}3L3 edges but only L3−2L^{3}-2L3−2 independent stabilizer constraints, leaving k=2L3+2k=2L^{3}+2k=2L3+2 logical degrees of freedom. We provide a minimum-weight perfect matching (MWPM) decoder adapted to the FCC geometry, demonstrate a 10x coding gain at p=0.001p=0.001p=0.001 (and 63x at p=0.0005p=0.0005p=0.0005), and discuss implications for fault-tolerant quantum computing on neutral-atom and photonic platforms.
A fault-tolerant quantum computer needs a memory code and a source of magic states. On a neutral-atom array the two can use different codes. This paper connects a [[15,1,3]] color-code factory, which has a transversal T gate, to the FCC sheet code, a [[3L³, 6L, L]] CSS memory at K = 4 connectivity. The connection is a transfer: a logical state is moved from a color-code block into one sheet-code logical by one-bit teleportation. We build the transfer circuit with a flagged joint Z̄ₐZ̄ᵦ measurement and simulate it at L = 4 under circuit-level depolarizing noise. The noise-free truth table holds for all four basis states. The circuit distance is 4 in the Z sector and 3 in the X sector. At p = 10⁻³ the transfer fails at 0.17% (Z) and 1.7% (X); the sheet memory alone fails at 0.07% and below 0.27% over the same rounds. When all noise on the color-code block is turned off, the X-sector excess disappears. The cost of the transfer is therefore the distance-3 factory block failing while it waits, not the interface. Two further results support the architecture. The lattice rotation R: (x, y, z) ↦ (y, z, x) maps each of the three sheets onto the next. We verify at L = 4, 6, 8 that R is a CSS isomorphism and that it permutes the 2L logicals of a sheet. The sheets are copies of one code with a pairing fixed by the lattice, and the standard transversal CNOT between copies of a CSS code applies to any two sheets. One CNOT layer on R-paired qubits gives 2L logical CNOTs. Under noise, decoding the two sheets separately fails on the target sheet, with an apparent threshold near 0.3%, because the CNOT copies the control sheet's residual errors onto the target. A feed-forward decoder that propagates the decoded control errors through R restores suppression with L, with a crossing near 0.8% against the memory's 1.1%; with exact knowledge of the control errors the memory value is recovered. We also count the atoms used by the cross-sheet triangle-ribbon merge of the companion surgery paper: 3L/2 against 2L for a surface-code routing channel. We state what is not new. Each sheet is L rotated toric codes, so the memory is the toric code's. A transversal CNOT between copies of a CSS code is available to any stacked code, including surface-code patches on atom arrays. The sheet code adds a lattice-fixed pairing and the ribbon merge, not a throughput advantage; the pairing is not spatially local, since the distance between R-paired qubits grows as about 0.47L.
Despite significant progress on quantum low-density parity-check (qLDPC) codes, building qLDPC processors that are high-rate, high-throughput, hardware-friendly, and fast-to-decode remains a challenge. We introduce mitten codes, a family of qLDPC processor codes of encoding rate 20%20\%20% and check weight 999, based on non-abelian groups. Their non-abelian structure evades distance bounds constraining abelian counterparts, allowing mitten codes to reach distance 181818 and beyond with just a few hundred data qubits. The logical operators of a mitten code are related by the group action, yielding a modular, low-overhead logical toolkit: full Clifford operations follow from bridging two reusable seed surgery gadgets or from a single fixed extractor. Furthermore, qLDPC processors based on mitten codes support high-rate surgery that executes many logical measurements in parallel, and parallel magic-state injection into all logical qubits at once. Under circuit-level noise, with our fast decoder, the [ ⁣[300,60,14] ⁣][\![300,60,14]\!][[300,60,14]] mitten code achieves, without extrapolation, a block logical error rate of ∼10−11{\sim}10^{-11}∼10−11 per round at 0.1%0.1\%0.1% physical error rate (PER), while the [ ⁣[975,195,≤24] ⁣][\![ 975,195,\leq 24 ]\!][[975,195,≤24]] code reaches ∼10−8{\sim}10^{-8}∼10−8 at 0.4%0.4\%0.4% PER. Decoding 151515 billion surgery experiments on the [ ⁣[540,108,18] ⁣][\![540,108,18]\!][[540,108,18]] code at 0.1%0.1\%0.1% PER, we observe only two logical failures, demonstrating a qLDPC processor capable of running ∼1010{\sim}10^{10}∼1010 logical operations. Our decoder is compatible with sub-millisecond average latency per logical cycle, sufficient for real-time decoding on neutral atom hardware. Discovered by an end-to-end design pipeline built on sQetch, a distance estimator orders of magnitude faster than existing tools, and mapping efficiently onto near-term neutral atom and superconducting hardware, mitten codes open a practical path toward fault-tolerant quantum computation.