The arXiv preprint (v1, May 2026) demonstrates dissipative preparation of approximate thermal states for kagome antiferromagnets using IBM hardware, reaching 79 system spins plus 60 environment qubits for a total of 139 qubits in the Ising case. Methodology relied on engineered dissipation via repeated two-qubit gate layers exceeding 1000 depth, with steady-state emergence observed at adjustable effective temperatures; classical statevector simulations benchmarked the protocol on up to 24-site lattices, showing circuit depth independent of system size and at most linear in inverse temperature. This preprint remains non-peer-reviewed. While the work correctly highlights escape from the QMC sign problem for the Heisenberg model, it underplays how IBM's heavy-hex lattice connectivity and calibration drift already constrain deeper circuits, a gap earlier experimental papers on variational thermal states (e.g., Nature Physics 19, 2023 on 2D Ising thermalization) had quantified through error-mitigation overhead. Synthesizing with Google Quantum AI's 2024 results on dissipative cooling in superconducting arrays reveals a shared pattern: dissipation stabilizes steady states faster than unitary methods but inherits hardware-specific decoherence spectra that theoretical proposals routinely omit. The 139-qubit demonstration therefore marks hardware progress that purely algorithmic quantum-simulation literature continues to overlook, implying near-term devices may soon map finite-temperature phase boundaries inaccessible to tensor-network or Monte Carlo routes.