AICCM real-Gamma production track

Experimental. The real-Gamma producer stays under AICCM2026DevAExperimentalWarning until the maintainer decides otherwise. This page is the plan and the ledger for getting it there: what “production ready” means for this line, the target architecture, and the increments that land it in small steps on main. It is updated as increments land.

1. Scope

The real-Gamma producer of the neutral fitted-torus Hamiltonian: run_ccm_rhf_direct, run_ccm_rks_direct, run_ccm_uhf_direct and run_ccm_uks_direct in vibeqc.periodic.ccm.direct, the neutral cderi fold and its block storage (neutral.py, block_cderi.py), the runner arm run_real_gamma_scf and run_periodic_job(method="aiccm", variant="real-gamma"). Its sibling is the neutral-Bloch producer (run_ccm_rhf_gdf, run_ccm_rks_gdf), the Bloch representation of the same Hamiltonian.

Out of scope: the union-and-weight four-centre lineage (aiccm2026dev-a construction routes) and chi-CCM (aiccm2026dev-b). Nothing here changes their Hamiltonians or labels.

2. What production ready means for this line

Each criterion is checkable; the increments in section 4 name the test that checks it.

1. One Hamiltonian, two representations. Real-Gamma and neutral-Bloch agree to 1e-10 Ha/cell (the Theorem-1 identity) on the ladder set: LiH, MgO, diamond and fcc Be at STO-3G and pob-DZVP-rev2, meshes (1,1,1) to (2,2,2), HF, PBE and PBE0, closed and open shell, with the RSGDF tail resolved at every N (#441, landed) and no parity hold.

2. Same-source evidence. Every result carries a source identity: fit source, auxiliary basis, kinetic-energy cutoff and tail, cell-list cutoff, exchange seam convention, storage mode and symmetry plan. A comparison across different sources is refused by identity, never averaged.

3. Evaluated state (#472). The terminal Fock, density, energy components, source and frame identities are frozen in a FockBuild. DIIS trial Focks are solver operators and never the reported physical operator. The .system manifest and the QVF vendor section carry the build identities.

4. Robust SCF. SAD, READ and Bloch-seeded initial densities on every direct driver and in the runner arm, a frontier-gap report and an internal stability check after convergence, so a second stationary point is found rather than reported (#442).

5. Cost. Block storage is the default for N_c > 1 with the dense fold as the reference; the XC image domain matches the Bloch side at the Bloch side’s cost; the fold cost is tracked (#258); symmetry reduction is opt-in and records its qualification evidence.

6. Interchangeable providers. The producer runs on the classic RSGDF fit (the libint/native FFT builder of the multi-k GDF driver) and on MANTA periodic GDF tiles. Each is a named source with its own Bloch control.

7. Promotion evidence. The identity tests move from T3 experimental to T2 rows with measured runtimes, the manual pages follow behaviour, and parity against PySCF and CRYSTAL23 is recorded in the AICCM-2026 set.

3. Target architecture

3.1 Fit-source provider boundary

Today the real-Gamma cderi is built by ccm_neutral_cderi_fold: per kept momentum transfer q, one shared-q batch of unit-cell Bloch-pair fits L_{k,k+q}[P,mu,nu] from vibeqc.aux_basis.build_lpq_bloch_native_fft_shared_q (the multi-k GDF driver’s classic RSGDF builder), folded to the torus or kept as block storage. The per-channel fits are the natural provider boundary, and since RG-7 it is explicit (vibeqc.periodic.ccm.fit_source):

fit source               bind(FoldContext) -> bound source with
                           channel_fits(qi, bras) -> {ai: L_{ka,ka+q}[P,mu,nu]}
                             in the canonical auxiliary frame, ket unfolded
                           supports_time_reversal, supports_star_plan
                           engine, transport, receipt()
ClassicRSGDFFitSource    the builder above (shared-q or per-pair batching)
DriverFitCacheSource     the multi-k driver's retained fit of the same operator
MANTAFitSource (RG-8)    vibeqc.manta.PeriodicGDFTiles per q over the mesh,
                         projected with the RI-Q builder's positive projector

ccm_neutral_cderi_fold and ccm_neutral_cderi_blocks take fit_source=; the fold keeps its orchestration (kept channels, time reversal, the star plan, the fold or the block storage) and applies time reversal and the star plan only to a source that declares them. CCMBlockCderi stores what it gets and contracts J and K exactly as before. The source receipt becomes part of cderi_info and of the SourceProvenance of section 3.3. MANTA kernels stay on their admitted native path (PeriodicGDFTiles.evaluate, resource budgets, receipts); the adapter copies tiles into the block storage it owns and never wraps a generic copying consumer around the native loop.

A MANTA-sourced torus is a different neutral source from the classic one (docs/design/manta_scientific_contract.md, section 6): its Bloch control is MANTA’s own whole-mesh RI build (PeriodicRIQBuilder at the full BvK mesh), not run_ccm_rhf_gdf. The Theorem-1 identity is proved per source.

3.2 Storage

Block storage (cderi_storage="block", commit cccadc6e) keeps 16 N_kept N_c n_aux n_mu^2 bytes against 8 n_aux n_mu^2 N_c^3 for the dense fold and evaluates J and K in the mesh representation on the circulant projection of the torus density. It is wired for RHF and RKS. The open-shell drivers fail closed on it. The track extends it to UHF and UKS (per-spin exchange through the same per-k factor) and then makes it the runner default for N_c > 1, keeping the dense fold as the reference switch and the tests/test_ccm_block_cderi.py identity at 1e-10 Ha/cell as the gate.

3.3 Fock build and evaluated state

The SCF loops (ri._rhf_loop, ri._rks_loop, the inline open-shell loops in direct.py) assemble F = h + J - 1/2 [K + xi_N S D S] (plus V_xc and the a_x fraction for KS) and the energy from the same D. The #472 contract is adopted without changing those formulas:

• a SourceProvenance per run (provider ccm-neutral-real-gamma, details from cderi_info, the one-electron fold’s lattice options, the exchange seam, the XC domain and grid);

• an OperatorFrame built from the supercell S and the canonical orthonormaliser X the loops already use;

• at the accepted state, FockBuild.from_dense with EnergyComponents (hcore, coulomb, exchange, exchange_q0_seam, xc) whose sum is the electronic energy the loop reports; one restricted channel at Gamma, or alpha and beta channels;

• DIIS trial Focks described through FockBuild.solver_operator and never stored on the result as fock;

• result.fock_build and FockBuild.output_fields() in the runner manifest and the QVF vendor section.

Acceptance follows the #472 pattern of the Gamma Ewald producers: same-density components, the energy derivative F = dE/dD by finite differences on a trial density, terminal-state ownership (mutating result arrays does not change the frozen build), and the output identities.

3.4 Symmetry

vibeqc.periodic.ccm.symmetry already supplies the cluster-invariant space-group subgroup, the supercell AO rotations, the atom-pair orbits and the fold’s k-pair star plan (cderi_symmetry, opt-in). The shared symmetry work supplies periodic_k_symmetry.star_operations and expand_k_matrices_to_full (the multi-k GDF exchange wedge) and the qualification contract of symmetry_shared: OperatorContract, QualificationEvidence and audit_equivariance, where a passed probe is not a production authorization.

The real-Gamma producer consumes these, never reimplements them:

• channel stars. Block storage keeps one representative per q star and transports the others through the star operation at contraction time (the fold’s star plan already reconstructs them at build time);

• wedge exchange. Kh(k) is built at wedge k only and transported across each star with the same transport the multi-k GDF wedge uses, so both producers share one implementation and one admission rule (the classic fit admits equal-lattice-shift operations; the pair-centred fit admits all);

• qualification. At build time the producer runs audit_equivariance on a nonsymmetric trial density (F[g.D] = g.F[D] for every admitted operation) and records the QualificationEvidence in cderi_info. The reduction is opt-in (symmetry=True or a CCMSymmetry), never a default, and refuses an operation the audit fails.

3.5 XC

Both producers evaluate the XC term with build_xc_periodic on the periodic Becke grid over the image domain of periodic_xc_domain.xc_lattice_opts (#465, #512; c9d32436 grew the real-Gamma domain to match). That domain made the real-Gamma LiH/STO-3G PBE SCF fourteen times slower. The track measures where the cost goes (AO-reach screening statistics against the dense counts) and recovers it inside the shared kernel, so the Bloch side gains the same; the identity to the Bloch side at 1e-10 Ha/cell is the gate.

4. Increments

Each row lands on its own in a day or two, keeps main working and is gated by the named test. Sizes: S under a day, M one to two days, L a week.

Id	Increment	Size	Depends on	Acceptance	Status
RG-1	XC cost after the grown domain: measure the kernel at 15 bohr and at the #465 bound on LiH/STO-3G PBE, then recover the cost without changing a number	M	none	E_xc unchanged to 1e-10 Ha at both domains; LiH PBE (1,1,1) real-Gamma SCF within 2x of its pre-#512 time; tests/test_ccm_xc_domain.py green; the #512 xfail in tests/test_ccm_block_cderi.py removed once the LiH PBE (2,2,2) row passes	RG-1a landed: the builders ride the kernel’s finite-torus regrouping (identity to rounding). RG-1b landed as an opt-in: a profile of the grouped build put about 80% of it in the torus AO sums, because the ball-shaped pair domain makes the rectangle cover degenerate to about one rectangle per image; xc_torus_complete=True on the KS drivers (kernel torus_complete=True) sums every pair of images reaching a batch through its torus residue instead. It is the domain-converged XC term, not the truncated one: E_xc at the 31.4-bohr bound and at 40 to 60 bohr agree to 7e-12 Ha on LiH (1,1,1), but on the H2 cube (2,1,1) the bound (14.0 bohr) truncates E_xc by 3e-10 Ha, which the complete density removes (it equals the ungrouped kernel at 30 bohr to 3e-13). LiH PBE (1,1,1): XC build 4.2 s against 75.9 s (pair graph) and 214.9 s (ungrouped), six threads; SCF 40 s against 444 s, 2.9e-11 Ha/cell apart, and 61 s for the pre-#512 15-bohr kernel. Because it changes the term at the bound’s truncation level (up to 1.5e-10 Ha/cell) while the Bloch producer keeps the pair graph, it is opt-in and recorded in the neutral-source identity; making it the default on both producers is a section 5 decision. The #512 xfail is removed: test_block_identities_rocksalt_222[LiH-pbe] passes at 1e-10 Ha/cell (30 min on six threads with the pair graph)
RG-2	Source provenance: a SourceProvenance record on every direct and Bloch result, in the manifest and QVF vendor section; cross-source Theorem-1 comparisons refused by identity	S	none	tests/test_ccm_neutral_source.py: both producers and both runner arms agree on the H2 cube (2,1,1); every Hamiltonian-selecting knob changes the identity; an injected array is refused; the direct-vs-GDF gate of tests/test_ccm_direct.py checks it	landed
RG-3	FockBuild for the closed-shell loops (_rhf_loop, _rks_loop), result.fock_build, output fields	M	RG-2	tests/test_ccm_fock_build.py on the H2 cube (2,1,1), HF and PBE: the terminal build is the reported state, its Fock is the derivative of an independent energy functional (2e-9), it owns its arrays, an iteration-capped run reports an evaluated state, the manifest carries the identities; energies unchanged to rounding	landed
RG-4	FockBuild for the open-shell loops (UHF, UKS)	M	RG-3	tests/test_ccm_fock_build.py open-shell rows on the Li doublet cube (2,1,1): the two-channel terminal build is the reported state, its Focks are the per-spin derivatives of an independent energy functional, an iteration-capped run reports an evaluated state, the UKS build carries E_xc, the manifest carries the identities; the UHF loop now tests convergence before DIIS	landed
RG-5	Block storage for UHF and UKS	M	none	tests/test_ccm_block_cderi_open_shell.py: identity to the dense fold at 1e-10 Ha/cell on the Li doublet cube (2,1,1) for UHF, PBE and PBE0; the drivers build block storage themselves; the Wigner-Seitz partition reaches them	landed
RG-6	Block storage as the runner default for N_c > 1; dense stays the reference switch	S	RG-5	runner parity rows; cderi_storage recorded in the manifest	the opt-in landed: run_periodic_job(method="aiccm", variant="real-gamma", aiccm_cderi_storage="block") (or "block-star") and run_real_gamma_scf(cderi_storage=...), recorded as cderi_storage_executed in the manifest and printed in the .out header; refused off real-gamma and with correlation=. On the H2 cube (2,2,1) the block and star runs equal the dense energy to 1e-10 Ha/cell with one neutral-source identity (tests/test_periodic_ccm_real_gamma_adapter.py). Open: the default itself, a maintainer decision (section 5)
RG-7	Fit-source boundary: ClassicRSGDFFitSource wrapping the shared-q and per-pair builders, DriverFitCacheSource wrapping the fit cache; the fold and the block storage consume a bound source	M	RG-2	tests/test_ccm_fit_source.py on the H2 cube (2,1,1) and (3,1,1): the default source is the historical fold bit for bit, a foreign source serving recorded fits reproduces the fold and the block storage exactly, the batchings and the driver cache are one operator, undeclared relations are refused	landed
RG-8a	MANTAFitSource over PeriodicGDFTiles and the RI-Q factor, on the domains of MANTA’s fitted RI HF, one tile source per kept channel	M	RG-7	tests/test_ccm_manta_fit_source.py on the H2 cube (2,1,1) and (3,1,1): block-storage J/K on MANTA fits equal MANTA’s whole-mesh PeriodicRIFockBuilder to 1e-16, the dense fold of the fits gives the same J/K, the source records its engine and channels, the SCF converges on it	landed
RG-8b	MANTA one-electron source (S, H and nuclear repulsion from MANTA’s finite supports) on the HF direct drivers (oneelectron=), so a MANTA-sourced torus is MANTA’s whole Hamiltonian	M	RG-8a	tests/test_ccm_manta_one_electron.py: the one-electron torus equals MANTA’s RI HF overlap/core Hamiltonian to 1e-14 and its nuclear repulsion to rounding; real-Gamma RHF on the H2 cube (2,1,1)/(3,1,1) and UHF on the Li doublet (1,1,1) equal MANTA’s RI HF energies to 1e-10 Ha/cell; the supercell Madelung constants agree to 7e-14. Open: at N_c > 1 a doublet cell is a supercell singlet in CCMSystem but per-cell spin in MANTA (42 mHa/cell apart on Li (2,1,1)); a maintainer decision	landed
RG-8c	MANTA-sourced KS: MANTA’s periodic XC quadrature as the XC source of the real-Gamma KS drivers	M	RG-8b	real-Gamma RKS/UKS on MANTA sources equals MANTA’s fitted KS jobs at the same mesh	landed: MANTAXCSource (xc_source= and oneelectron= on the KS drivers); MANTA’s KS domains have the translations and Gamma vectors of its RI HF domains, so the RG-8a/8b sources on the twin HF options serve the KS run. tests/test_ccm_manta_xc_source.py: RKS PBE on the H2 cube (1,1,1)/(2,1,1)/(3,1,1) equals run_periodic_ri_ks_from_cutoffs to 7e-16 Ha/cell, UKS PBE on the Li doublet (1,1,1) to 1.4e-14; the torus potential is the derivative of the torus energy. Semilocal only
RG-9	Guesses and stability: SAD and Bloch-seeded densities on the direct drivers and the runner arm; frontier-gap report; RHF-to-RHF stability check	M	RG-3	MgO pob-DZVP-rev2 RHF (1,1,1) lands on the Bloch solution (#442); the HCORE seed on the same cell is reported as a higher stationary point	source landed: an explicit SAD on the four direct drivers, run_real_gamma_scf and the runner (AUTO stays HCORE, through periodic_guess_capabilities(route, requested=...)); the Bloch seed is READ of ccm_reference_from_bloch(...).density; frontier_gap on the convention record and frontier_gap_ha in the manifest; stability_check / follow_instability on the RHF driver (vibeqc.periodic.ccm.stability: singlet and triplet sectors by Davidson, translation-invariant following). tests/test_ccm_rhf_stability.py: the Davidson eigenvalues equal the explicit Hessian, the energy curvature along the mode is four times the eigenvalue, and the uniform equidistant carbon chain at (5,1,1) is a singlet saddle (-0.081 Ha) that the following lowers by 7.8 mHa/cell. Found and fixed on the way: the HF loops reported the orbitals of the last DIIS trial Fock. Acceptance met on the untailed Hamiltonian of #442 (MgO pob-DZVP-rev2 (1,1,1), classic fit at 200 Ha, no RSGDF tail): the Bloch control and the HCORE real-Gamma run reproduce #442’s -274.8696749719 and -274.6156706666 Ha/cell exactly; the HCORE state is a singlet saddle (-0.126 Ha); following it (one step), SAD and the Bloch seed all land on -274.8696749719. The Bloch state is singlet-stable (+0.139 Ha) with a weak triplet instability (-0.0074 Ha). Converging the Bloch SCF to 1e-12/1e-10 lets ccm_reference_from_bloch accept at its 1e-8 density tolerance (#442’s open residual). Open: the tailed variant, whose tail (54,492 Ha, about 7.7e7 plane waves) puts the fit in the #258 cost class
RG-10	Channel-star reduction of the block storage with audit_equivariance evidence	M	RG-7	reduced and unreduced J and K agree to 1e-12 on random nonsymmetric densities for diamond Fd-3m and MgO (2,2,2); SCF energy unchanged to 1e-10 Ha/cell; evidence in cderi_info	landed: cderi_storage="block-star" keeps the star plan’s representative fits and a recipe (operation, two orbital maps) per member; the unitary auxiliary map cancels in the exchange sum, so the reduction is exact for any density. Evidence: cderi_info["star_storage"] (accounting and a build-time recipe check, 3e-16 on the H2 cube). tests/test_ccm_block_star.py: on compact H2 (2,2,2), 42/64 fits kept, J/K on random nonsymmetric densities equal the materialised storage to 3e-17 and the fully built one to 1.1e-15, RHF/UHF energies through the drivers equal the dense path; the diamond and MgO (2,2,2) rows are slow. Acceptance met on both cells: diamond Fd-3m (2,2,2) passes its slow row; MgO Fm-3m (2,2,2), whose pytest row exceeds 3 h single-threaded (the unreduced classic fold is in the #258 cost class), was run with the same checks on four threads: J and K against the fully built storage 8.3e-13 and 8.8e-13 (K scale 8.2), recipe check 2.2e-15, 24 of 64 fits stored (reduction 2.67), RHF energy 3.4e-12 Ha/cell from the built storage; the three folds took 2340, 2145 and 4573 s. The evidence is the identity check rather than audit_equivariance, which tests a different property (equivariance of one operator)
RG-11	Wedge exchange on the block storage through the shared k-star transport	M	RG-10	as RG-10, plus the measured flop and time reduction recorded on this page	landed: exchange_wedge=True builds K on one bra per k-star of the plan (read off its q=0 channel) and transports it with Om = P_ao diag(e^{i (Rk).S}), the periodic_k_symmetry convention; every call measures the density’s distance from the plan’s symmetry and builds K in full when it is off (cderi_info["exchange_wedge"]). tests/test_ccm_block_wedge.py: H2 (2,2,2) builds 6/8 bras and equals the full K on the SCF density to 1e-13; RHF and UHF through the wedge equal the full build with no fallback. Diamond (2,2,2) STO-3G: 4/8 bras, so the bra work of K halves; measured K build 3 ms full against 2 ms wedge on a loaded host (the per-member transport is n_mu^3). The reduction grows with the star size, up to the cluster group order for a general k
RG-12	Fold cost (#258): profile ao_pair_fourier_transform_bloch on diamond and MgO (2,2,1) at a 25-bohr list and recover the regression	L	none	the three tests/test_ccm_symmetry.py fixtures finish inside their 1800 s budget	partly landed by another route: RangeSeparatedFitSource puts the fold on the driver’s default SR/LR fit, which needs no tail. On MgO pob-DZVP-rev2 (1,1,1), whose classic tail (54,492 Ha) kept the Bloch fit in ao_pair_fourier_transform_bloch_multi for over four hours, the SR control takes 66 s and the SR fold 53 s, and the two producers agree to 5e-11 Ha/cell (tests/test_ccm_range_separated_source.py). The SR source admits the star plan and time reversal (961243fd; its canonical auxiliary frame is covariant): on diamond (2,2,2) the SR fold takes 399 s with the plan and 385 s as star storage against 996 s unreduced (four threads), 24 of 64 fits stored, J and K within 4.8e-13 and 5.7e-12 of the unreduced fold at a K scale of 4, the level of the SR builder’s integral screen (1.25e-11), whose decisions are not exactly symmetric; on the H2 cube (2,2,2) they agree to 3e-15. The per-channel cost itself is #543: each SR channel recomputes its real-space integrals per q and its reciprocal pair transforms per ket, 50 to 140 times PySCF’s time at (2,2,2). The fold’s SR source now shares the q-independent real-space sources across its channels through the multi-q builder of #543: diamond (2,2,2) unreduced 161.6 s against 833.8 s per channel (four threads), bitwise the same fits. Open: the rest of #543 (owned by the GDF builder), the classic kernel’s own cost, and making SR the CCM default on both producers (a maintainer decision, see section 5)
RG-14	Explicit spin-state contract: one spin_state keyword on CCMSystem, the direct and Bloch producers, both runner arms and run_periodic_job(aiccm_spin_state=); the sector resolved before the shell-aware dispatch; the executed counts on every result, in the manifest and the QVF record; a comparison refused by sector as well as by operator	M	RG-2, RG-4	tests/test_ccm_spin_state.py: the legacy mismatch reproduced with its counts and refused by sector; the per-cell doublet on (2,1,1), (3,1,1) (Li) and (2,2,1), (2,2,2) (H cube) equal on both producers to 1e-10 Ha/cell with the Bloch projectors folded onto the torus; UKS PBE at (2,1,1); impossible requests, restricted drivers on an open sector, a foreign-sector READ and a non-uniform total on the Bloch producer refused; manifest round trip	landed: vibeqc.periodic.ccm.spin_state (CCMSpinState, resolve_ccm_spin_state, assert_compatible_spin_state, assert_comparable_neutral_results). "legacy" (default) changes no number: real-Gamma pairs a doublet cell by supercell parity, the Bloch producer replicates the cell’s counts, and the mismatch (Li STO-3G (2,1,1): 3/3 against 4/2, 4.83 mHa/cell on the classic fit; the 42 mHa/cell of RG-8b was MANTA’s def2-SVP-JK source) is now recorded and refused instead of silent. "per-cell" and an explicit total multiplicity select one sector on both producers; the Bloch producer refuses a sector that is not N_c copies of one pair (Theorem 2 of the theory notes holds for block-circulant per-spin densities with Tr[D_sigma S] = N_c n_sigma, which is also the state assumption behind the open-shell representation comparison; Paper 1 states Theorem 1 for closed shells). Found on the way: the Li per-cell quintet at (2,2,1) has a degenerate alpha frontier that the multi-k driver occupies fractionally (an ensemble the real-Gamma determinant cannot hold, 11 mHa/cell apart, refused by ccm_reference_from_bloch), so a same-sector comparison also needs integer occupations; the H cube doublet carries the multidimensional rows. A total multiplicity fixes n_alpha - n_beta only; antiferromagnetic or other broken-symmetry arrangements need a per-cell seed the producers do not take and are not claimed. Scientific acceptance through vq against PySCF: see the #519 note of this increment
RG-15	Occupation identity: an OccupationRecord on every result of both producers (integer or ensemble, largest departure, per-k counts, frontier gap, degenerate flag), in both convention records and the manifest (aiccm_occupation); assert_comparable_neutral_results refuses an ensemble by name before any fold	S	RG-14	tests/test_ccm_spin_state.py: the Li per-cell quintet at (2,2,1) is recorded as an ensemble (departure 0.5) and refused; the gapped rows carry equal frontier gaps on both producers; record round trip and manifest	landed. The fix itself, a fixed per-k integer fill, belongs to the shared multi-k driver (pbc-gdf): until it exists a degenerate-frontier torus is a real-Gamma-only row with its recorded frontier
RG-13	Promotion evidence: ladder parity against PySCF and CRYSTAL23 recorded in studies/aiccm-2026; the identity tests moved to T2 rows with measured runtimes; manual pages current	M	RG-1 to RG-6	manifest rows with measured_runtime_s; the AICCM-2026 tables	landed except the move to T2, a section 5 decision. Measured runtimes: the 18 identity files under the AICCM lanes’ -m "not slow" through scripts/test_gate/run_full_suite.py (3 jobs, source be62b6aa); 17 pass, 10 to 398 s each and 34 min in sum, and their manifest rows carry measured_runtime_s (the tier stays T3, see section 5). tests/test_ccm_rks_direct.py did not finish: the per-test 1800 s cap stopped test_rks_direct_matches_gdf_pure_gamma_post_fix in the multi-k XC build of LiH PBE (1,1,1), the RG-1 cost. The manual documents the real-Gamma Python controls (storage, wedge, fit, one-electron and XC sources, guesses and stability). PySCF ladder (studies/aiccm-2026/ladder_pyscf.py, README section RG-13): the SR fit equals PySCF KRHF rs_density_fit with the transported shells to 0.0 to 18 uHa/cell on diamond, LiH, MgO, fcc Be STO-3G (1,1,1) to (2,2,2) and MgO pob-DZVP-rev2 (1,1,1), PySCF’s own GDF/RSGDF spread; the classic fit sits 32 to 215 uHa/cell above PySCF GDF. The SR route costs 50 to 140 times PySCF’s time at (2,2,2) (#543). CRYSTAL23 (SHRINK 8, TOLINTEG 8 8 8 8 16, the transported STO-3G; SHRINK 12 moves diamond by 6 uHa): from (4,4,4) on the PySCF sequences, which the real-Gamma producer equals rung by rung, approach it from below and sit 2.5, 1.0 and 2.5 mHa/cell away at the largest mesh run (diamond (6,6,6), LiH (6,6,6), MgO (5,5,5)); the torus exchange of the two-atom cells converges slowly, so no extrapolated limit is quoted. Open: the RG-10 MgO (2,2,2) row outside pytest; promotion to T2 (section 5)

5. Maintainer decisions

These are not taken by the track; they are listed so they can be taken.

• Removing AICCM2026DevAExperimentalWarning from the real-Gamma producer.

• Making block storage the default (RG-6) and, later, any default symmetry reduction (the track keeps symmetry opt-in).

• Whether a MANTA-sourced real-Gamma run is exposed through the runner under its own selector, and under which name.

• Promoting the identity tests to T2 rows (RG-13).

• The spin convention of a doublet cell on a torus of more than one cell (RG-8b, RG-14): the default spin_state="legacy" keeps the supercell parity pairing on the real-Gamma producer and the per-cell replication on the Bloch producer and MANTA; "per-cell" makes every producer run the uniform aligned sector. Whether "per-cell" becomes the default.

• Making SAD the AUTO default of the real-Gamma variant, and running the RHF stability check by default as the molecular drivers do (RG-9).

• Making the complete torus XC density (RG-1) the default of the real-Gamma KS drivers, together with the Bloch producer’s XC kernel so that the two stay one operator; it is the domain-converged XC term and much cheaper.

• Moving both neutral producers from the classic fit to the SR/LR fit, the multi-k driver’s default 3-D source (RG-12): the pin to the classic fit in ri._pin_classic_fit_source lifts on both halves together.

6. Coordination

• #512 (real-Gamma and Bloch KS split): closed by c9d32436 (option 1, real-Gamma grows its XC domain). codex/qc512-ccm-xc-domain is superseded; its remaining extras are already on main. The dormant xfail in tests/test_ccm_block_cderi.py is removed under RG-1.

• #513 (stale diamond assertion): codex/qc513-classic-fit-assertion.

• #404 (classic-fit gradient): landed as 4715634f.

• #441 (tail at every N): landed, changelog d6f60fb6.

• #442 (READ densities): landed as 6a7a09fd; the SAD, Bloch-seeded and stability parts are RG-9.

• #472 (unified Fock): the landed vibeqc.fock_state API is consumed as is; codex/fock-rhf-evaluated-state and codex/unified-operator-contract continue on the Gamma producers. The CCM adoption lives under periodic/ccm/ only.

• Shared symmetry (the codex/symmetry-* merges): consumed through symmetry_shared, periodic_k_symmetry and ccm/symmetry.py; no change to those modules from this track. Requests go on the tracker.

• MANTA: consumed through vibeqc.manta (PeriodicGDFTiles, PeriodicRIQBuilder) and the receipts they carry. A binding the adapter needs and does not find is requested on the tracker, not added here.