40-ion Ba-133 qLDPC testbed (IonQ: a Tempo engineering test system)
40-ion Ba-133 qLDPC testbed (IonQ: a Tempo engineering test system)
Register profile — the machine's variant of its architecture
qubit type: Ba-133 hyperfine
gate mechanism: laser Molmer-Sorensen
connectivity: all-to-all through the chain's motional bus
control: room-temperature electronics + free-space laser addressing
control placement: room-temp + free-space optics
readout: fluorescence
gate evidence: no published figure
decoder mode: offline — syndromes decoded after the run
codes: generalized_bicycle, qcga, rotated_surface, stab_16_4_4, stab_4_2_2, toric
flags: code params inferred, post-selection dominates the result
Technologies by layer — the machine's cell per layer
| 1 Qubit carrier | Trapped atomic ion · primary ▣ Fig. 1 (ion trap, Raman beams steered by AODs) ✅ 40 ¹³³Ba⁺ ions in one stationary chain; physical coherence 1.1±0.3 s; lab-only testbed (Jun 2026) |
| 2 Encoding | Hyperfine / clock-state qubit · primary ▤ §II (Mid-circuit measurement) 🔎 ¹³³Ba⁺ ground-state hyperfine qubit; family-determined, no machine-specific encoding source |
| 3 Gate mechanism | Mølmer–Sørensen / light-shift laser gate · primary ▣ Fig. 1 (counter-propagating Raman gates); Figs. 4–5 (MS gate DRB); App. C ✅ laser Mølmer–Sørensen entangling gates across the 40-ion chain |
| 4 Connectivity / transport | Ion-chain motional bus (all-to-all in chain) · primary ▣ Fig. 1 (Tanner graphs of required 2Q connectivity); App. F Qubit mapping ✅ all-to-all through the chain's motional bus (degree 39); eight codes in three families run with no hardware reconfiguration (nine in the June version) |
| 5 Control | Free-space laser addressing of ions (AOM/AOD beams) · primary ▣ Fig. 1 (Raman beams steered with AODs); Fig. 10 (gate/MCM schedule) ✅ room-temperature electronics + free-space laser addressing |
| 6 Readout | Fluorescence state detection (ions, defects) · primary ▣ Fig. 1 (OMG-based MCM sequence); §II; Fig. 6 (leakage detection/rejection) ✅ fluorescence readout with leakage detection; ~75 % post-selection rejection for the larger codes |
| 7 Code | Non-local qLDPC codes (bivariate bicycle, 'gross') · primary ▣ Fig. 2 (BB5/GB4 survival vs syndrome cycles); §III; App. A Code definitions ✅ eight codes in three families on one device (nine in the June version); break-even within error bars (logical 3.95±0.68 s in one code vs physical 3.84±0.48 s = 3 T₂*, arXiv version of 14 Sep 2026; the June version gave 3.3±0.9 s); BB5 [[18,4,3]] 4× X / 9× Z below the prior superconducting BB run |
| 8 Decoder | — (Atlas gap: ∅G-beam) · primary ▤ §III (Quantum memory experiments) 🔎 beam-search decoder (Delfosse) for the qLDPC codes, run offline; no Atlas node for search decoders |
| 9 Interconnect | |
| 10 Manufacturing | Ion-trap fabrication (surface-electrode chips; machined 3D blade traps) · primary ◌ 🔎 surface-electrode ion-trap microfabrication; family-determined, no machine-specific fab source |
Standard records
acceptance fraction → 0.25 · typical for the larger codes (~75% rejection)
Jun 2026 · source
T2 coherence (echo/DD) → 4.0 s · best (logical lifetime, one code per the abstract)
physical baseline on the same device 3.84 ± 0.48 s (= 3 T₂*; arXiv version of 14 Sep 2026; 3.3 ± 0.9 s in June) · Sep 2026 · source
evidence: ✅ 6 / 10
Pictures
Every picture carries the story of what it shows and its credit line.
