Complete, challengeable quantum projects from students, hobbyists, independent researchers, and small teams. Every entry links the official work to a hardware implementation, a classical competitor, and an explicit claim boundary.
What counts here
Local practical advantage means that a measured quantum workflow reached a useful answer faster than a named classical workflow for the same stated task on the resources actually available to the project. It is not proof against every classical algorithm or supercomputer.
A stronger classical result is not a problem for this list. It is a successful challenge. The entry and its classification should change when the evidence changes.
One of the measured outcomes behind the list: the 120-qubit Fermi-Hubbard charge-density profile. The sector-plus-readout route is diagnostic because postselection discarded 98.9 percent of shots.
Local time-to-answer separation
Fermi-Hubbard dynamics on 120 qubits
A 60-site Fermi-Hubbard hardware workflow produced local charge, spin, and double-occupancy observables and was compared with local MPS and observable-specific Majorana calculations.
Scale
120 qubits / 60 sites
Backend
IBM Kingston through Q-CTRL Fire Opal
Primary timing
33.148928 s
Timing scope
Estimated main plus readout circuit execution; excludes full human and cloud workflow time
Measured comparison. The quantum execution proxy was 272.50x shorter than the local chi=256 MPS wall time for this declared instance.
Quantum result. The hardware produced a full 120-qubit observable profile; raw mean double occupancy was 0.22862549 and readout-corrected mean double occupancy was 0.23067722.
Classical baselines
Method
Wall time
Status
Local quimb MPS with maximum bond dimension chi=256
9,033 s
not fully converged; maximum bond reached the requested cap
Local Majorana propagation with cutoff 2
20.96 s
faster than the quantum proxy but visibly inaccurate
Local Majorana propagation with cutoff 4
1,153.51 s
close to the chi=256 value for this selected observable
Non-Abelian SU(2) hadron dynamics on 120 active qubits
A Loop-String-Hadron implementation follows a differential hadron signal on a 60-site lattice and compares the quantum route with local circuit-MPS checks and published tensor-network and Pauli-propagation baselines.
Scale
120 qubits / 60 sites
Backend
IBM hardware through Q-CTRL Fire Opal
Primary timing
1.425408 s
Timing scope
Local hardware time including readout circuits; excludes queue, API, compilation, and local analysis
Measured comparison. Both the local circuit checks and the paper-native baselines show a substantial runtime separation under their declared timing definitions.
Quantum result. The local hardware route produced charge-sector and differential-observable data for the 120-qubit circuit family.
A tracker-compatible 80-qubit extension estimates an Operator Loschmidt Echo from finite computational-basis samples and compares the complete mitigated hardware action with a bounded tracker-linked BP-TN calculation.
Scale
80 qubits
Backend
IBM Kingston through Q-CTRL Fire Opal
Primary timing
328 s
Timing scope
Complete Fire Opal action wall time for all sixteen mitigated circuits
Measured comparison. The incomplete bond-dimension-64 classical delta half alone exceeded the complete Fire Opal action by more than 2.75x on this machine.
Quantum result. The measured delta/delta0 OLE ratio was 0.74028847 +/- 0.01663657; all eight sample ratios were positive.
Classical baselines
Method
Wall time
Status
Tracker-linked Heisenberg BP-TN at bond dimension 16
365.14 s
not converged; apparent ratio is not a valid physical estimate
Tracker-linked Heisenberg BP-TN delta half at bond dimension 32
342.42 s
not converged; value shifted by 86 percent from bond dimension 16
Tracker-linked Heisenberg BP-TN delta half at bond dimension 64
A complete 70-data-qubit non-Clifford circuit was sampled on IBM hardware, alongside an independent 70+8-qubit stabilizer-verification workflow and local classical scaling studies.
Scale
70 qubits
Backend
IBM Kingston through Q-CTRL Fire Opal and IBM Runtime
Primary timing
19 s
Timing scope
Provider quantum-seconds for 256 samples from the complete 70-data-qubit circuit; excludes queue and local workflow time
Measured comparison. Hardware returned 256 samples in 19 quantum-seconds, while a local Aer fit projects about 6.89 million years for one 70-qubit sample; sample counts and output quality are not matched.
Quantum result. The complete circuit returned 256 samples. The separate checked dataset retained 4,519 of 184,320 shots and gave a graph-state-prefix point estimate of 0.01217; its predeclared one-sided 95 percent lower-bound test failed. The original restricted-access IBM Boston execution reported substantially stronger effective performance than this independently accessible Kingston reproduction.
Classical baselines
Method
Wall time
Status
Local Qiskit Aer extended-stabilizer fit evaluated at 70 qubits
217,512,854,796,362.625 s
extrapolated; not measured at 70 qubits and not quality matched
Local ITensorMPS at maximum bond dimension 64
205.36 s
completed but strongly truncated and not converged
Local exact MPS anchor at 14 induced qubits
3.23 s
exact small-width validation; not a 70-qubit baseline
The Tracker result used restricted access to IBM Boston, whereas this independent reproduction used the available IBM Kingston route; backend access, physical mapping, and calibration window are therefore not matched.
Boston produced a substantially stronger workload-level result, but its historical calibration and complete raw fidelity-analysis record are not public, so the result does not establish that Boston was universally better hardware than Kingston.
The 70-qubit classical runtime is extrapolated from measurements ending at 12 qubits, not measured at full width.
The quantum samples have no validated full-distribution fidelity, and the separate predeclared 95 percent stabilizer test failed.
The post-hoc 75 percent lower bound is an exploratory sensitivity result, not 75 percent fidelity and not evidence of quantum advantage.
QOS-inspired PBMC68k feature generation on 60 qubits
A frozen 60-qubit QOS-inspired feature map generated 627 measured features for real PBMC68k cells on IBM Fez, reached the strongest held-out point score, and completed far sooner than the bounded local MPS attempt for the same specified feature target.
Scale
60 qubits
Backend
IBM Fez through Q-CTRL Fire Opal
Primary timing
26 s
Timing scope
Quantum-seconds reported by the Fire Opal dashboard for action 2335848; excludes orchestration, queueing, retrieval, data preparation, and classifier training
Measured comparison. Hardware generated the complete 60-qubit feature result in 26 quantum-seconds while local MPS remained incomplete after 2,577 seconds: a kernel-time lower bound greater than 99.1x; the complete Fire Opal route retained a lower bound greater than 5.0x.
Quantum result. Held-out balanced accuracy was 0.53125 (17/32), compared with 0.50000 (16/32) for the predeclared linear baseline and 0.43750 (14/32) for RBF. The exact McNemar p-value against linear was 1.0 and the paired-bootstrap 95 percent interval was -0.1875 to 0.25.
Classical baselines
Method
Wall time
Status
Local MPS convergence ladder for the same 60-qubit circuit and 627-feature target
2,577 s
stopped without a converged feature result
Training-only-selected linear SVC on classically prepared gene data
not available
completed; 0.50000 balanced accuracy (16/32)
Training-only-selected RBF SVC on classically prepared gene data
The MPS route did not converge, so the same feature target was specified but numerical feature equality at a matched error tolerance was not established.
The 26 quantum-seconds value was read from the Fire Opal dashboard; the archived get_result payload omitted the quantum-seconds field.
The greater-than-99.1x ratio compares QPU-only dashboard time with local MPS wall time; the broader submission-to-retrieval comparison is a lower bound greater than 5.02x.
The inexpensive classical linear and RBF classifiers do not require simulation of the 60-qubit feature map, so this is not an end-to-end speedup over ordinary classical machine learning.
The held-out test contains only 32 cells; the one-cell hardware lead is not statistically significant and does not establish general predictive advantage.
This is a local result under declared hardware and classical resources, not a claim against every tensor-network method, compute platform, or future implementation.
Copy the entry template, add the official source and full implementation, state both timing scopes, and preserve every convergence or accuracy limitation. Classical challenges are first-class contributions.
