QMBL lists 3 published ground-state energies for the spin-1/2 Heisenberg antiferromagnet (chain, 16 sites, periodic boundaries). The exact energy is -0.446393522538548, by ED in QMBL check, checks/heisenberg-chain-16-ed/; and the best variational energy is -0.446383(5), by VMC (ViT) in Shokry et al. (2025). Energies as E/N (S.S).
Every published energy
Grouped by what the number is. The exact energy comes first where there is one; in every other group the lowest unflagged energy is its best value, and each row's distance is to that best (rules §6).
Numerically exact (1)
Exact diagonalization, an exact solution, or sign-problem-free QMC where that is established. An exact energy is shown first, before the best of each other class. QMC is exact only within its error bar, which it must state, and reads exact (stochastic); an exact diagonalization outranks it. For a diagonalization resolved by symmetry sector only the ground state's sector is listed; the lowest energies of the other sectors are the instance's spectrum, in its JSON, and are not shown as the exact energy.
Strict variational upper bounds (2)
Energies that a variational state reached (VMC, DMRG and MPS at finite bond dimension, PEPS, NQS, VQE). Each is an upper bound on the ground-state energy.
How these numbers were read
Where each number below was read, and how it was checked.
-0.446393522538548, ED
Checked 2026-11-11 · ed_chain.py (code A: scipy.sparse Hamiltonian with ARPACK eigsh, true residual 2.4e-14; code B: matrix-free Lanczos with full reorthogonalisation, residual 8.1e-15), S^z = 0 space of 12,870 states, no spatial symmetry, numpy 2.5.4 and scipy 1.18.1 on an Apple M5 laptop, 2026-11-11; results in checks/heisenberg-chain-16-ed/results.txt
Reported as. -28.56918544246714 (code A) | -28.56918544246712 (code B) | -28.5691854424671194555711899928 (Bethe ansatz, checks/bethe-heisenberg-chain/bethe_xxx.py 16 P 30)
Pauli total; E/N = -0.446393523 in S.S units (S.S total -7.14229636062). H = sum_i S_i.S_{i+1} on the periodic chain (16 bonds). The S^z = 0 space holds a member of every spin multiplet, so its lowest level is the ground state. The same script reproduces the exact row of Heisenberg/chain_20_P to 6e-14. No published exact energy is carried for this instance; carried to 13 decimals, the agreement of the three calculations.
-0.446383(5), VMC (ViT)
Checked 2026-11-11 · source text of arXiv:2510.08430 read locally (arXiv HTML or pypdf layout text); value copied from the harvested cell and the text, no LLM transcription of numbers; single reading (2026-11-11)
Reported as. -28.5685(3)
Table 2, row 'Block-layer QGT', column Energy E (the Vision Transformer with a two-layer encoder of the chain benchmark, lines 130-131, optimised with the QGT block-diagonal by network layer); sources/2510.08430.txt line 193; version read: arXiv text (sources/2510.08430.txt), checked against the arXiv v2 PDF; no journal version (arXiv metadata and Crossref, 2026-11-11) | The text writes H = sum_i S_i.S_{i+1} with S_i = σ_i / 2 on the periodic chain of L = 16 sites (lines 126-128), but Table 2's energies are in the Pauli form: the exact ground state of that H is -7.142296 (QMBL's diagonalization and Bethe ansatz, checks/heisenberg-chain-16-ed/), and -28.5685(3) read with S = σ/2 would lie four times below it, impossible for a variational energy. Read as the Pauli total (4 x S.S, RULES.md 5), the two rows lie 6.9e-4 and 1.8e-3 above the exact Pauli total -28.569185, at 2.3 and 3.0 printed sigma. Stored as printed, the Pauli total; E/N in S.S units is the printed value / 64. | variance reported as σ² = 0.008 (Table 2, column Energy Variance), not stored: convention not verified
-0.446366(9), VMC (ViT)
Checked 2026-11-11 · source text of arXiv:2510.08430 read locally (arXiv HTML or pypdf layout text); value copied from the harvested cell and the text, no LLM transcription of numbers; single reading (2026-11-11)
Reported as. -28.5674(6)
Table 2, row 'Full QGT', column Energy E (the same Vision Transformer optimised with the full sampled QGT, 'full SR', line 131); sources/2510.08430.txt line 194; version read: arXiv text (sources/2510.08430.txt), checked against the arXiv v2 PDF; no journal version (arXiv metadata and Crossref, 2026-11-11) | The text writes H = sum_i S_i.S_{i+1} with S_i = σ_i / 2 on the periodic chain of L = 16 sites (lines 126-128), but Table 2's energies are in the Pauli form: the exact ground state of that H is -7.142296 (QMBL's diagonalization and Bethe ansatz, checks/heisenberg-chain-16-ed/), and -28.5685(3) read with S = σ/2 would lie four times below it, impossible for a variational energy. Read as the Pauli total (4 x S.S, RULES.md 5), the two rows lie 6.9e-4 and 1.8e-3 above the exact Pauli total -28.569185, at 2.3 and 3.0 printed sigma. Stored as printed, the Pauli total; E/N in S.S units is the printed value / 64. | variance reported as σ² = 0.026 (Table 2, column Energy Variance), not stored: convention not verified