GitHub

Changelog

All notable changes to this project will be documented in this file.

The format is based on Keep a Changelog, and this project adheres to Semantic Versioning.

Guidelines for updating this changelog

When making changes to this project, please update the "Unreleased" section with your changes under the appropriate category:

  • Added for new features.
  • Changed for changes in existing functionality.
  • Deprecated for soon-to-be removed features.
  • Removed for now removed features.
  • Fixed for any bug fixes.
  • Performance for performance improvements.

When releasing a new version, move the "Unreleased" changes to a new version section with the release date.

Unreleased

Added

Changed

Deprecated

Removed

Fixed

Performance

0.8.0 - 2026-06-05

Added

  • FullPullback and TruncPullback to specify pullbacks using a full untruncated or only truncated decomposition respectively (#382)
  • FixedPointGradient type, containing a specific solver for solving the fixed-point gradient linear problem (#386)
  • CTMRG benchmark suite (#384)
  • Rotation of LocalCircuit (#349)
  • LocalCircuit operator type, alongside refactor of LocalOperator operator types (#347)
  • N-site Simple Update (#339)
  • C4v CTMRG support, including QR-CTMRG variant (#321, #329)
  • Belief propagation gauge fixing, with bipartite specialization and support for purified iPEPO (#223, #318, #319)

Changed

  • Bump VectorInterface compat to include v0.6 (#389)
  • Make checks on rotation and reflection symmetry when using C4v CTMRG into warnings (#385)
  • Allow unsorted site indices in LocalCircuit (#387)
  • Periodic indexing improvements (#377)
  • Rename algorithm symbols (#376)
  • Update decomposition handling (#364)
  • Return info.converged flag and info.convergence_error in leading_boundary named tuple (#365)
  • Improve Vidal gauge conversion and CTM bond env for PEPO (#348)
  • Rename bondenv_fu to bondenv_ctm (#343)
  • Refactor simple update / reorganize trotter and apply_gate code (#338, #346). Each term of the input Hamiltonian LocalOperator is now exponentiated individually when constructing the Trotter gates in the resulting LocalCircuit.
  • Move info.truncation_error and info.condition_number into the info.contraction_metrics named tuple for leading_boundary and fixedpoint (#337)
  • fixedpoint improvements; relax env₀ type restriction in select_algorithm (#337, #345)
  • Use MatrixAlgebraKit SVD pullbacks (#335)
  • Bump TensorKit compat to 0.16 (#314)
  • Make SUWeight axis order definite (#315)
  • Bump OptimKit.jl compatibility to v0.4 (#307)
  • Improve bond truncation algorithms (#303). In particular, the 3-leg reduced bond tensors now follows the leg order convention of an MPS.

Removed

  • FullEighPullback, TruncEighPullback, FullSVDPullback, TruncSVDPullback and QRPullback types, in favor of more generic pullback specification across algorithms (#382)
  • LinSolver and EigSolver types, and iterscheme keyword for specifying the style of and solver for fixed-point differentiation (#386)
  • :diffgauge differentiation mode, also for eigh-based C4v CTMRG (#334, #370)
  • Unused implementations of LinearAlgebra methods for CTRMGEnv and InfinitePEPS (#307)

Fixed

  • Fix undefined dir variable in error message in TensorMap(::EnlargedCorner) (#392)
  • Ensure MPSKit overloads accept all keywords to intercept all dispatches (#374)
  • FixedSpaceTruncation for simple update (#360)
  • Stack overflow in renormalize_southwest_corner (#344)
  • Broken gradients for C4v eigh-CTMRG (#333)
  • CTMRG contraction inconsistencies (#327)
  • Typo in SimpleUpdate info (#313)
  • Typo in edge_transfer_right (#310)
  • MPS cluster truncation with standard virtual arrows (#309)
  • Small fixes for InfiniteSquareNetwork and InfinitePartitionFunction (#306)
  • Zero gate returned by get_gateterm (#300)

Performance

  • Improve simple update performance (#361)
  • Improve efficiency of bond truncation (#366)

0.7.0 - 2025-11-17

Added

  • Real time and finite-temperature evolution functionality for simple update
  • Correlator for mixed state InfinitePEPO
  • SUWeight to CTMRGEnv conversion (as InfinitePEPS environment)
  • Simple update for PEPO (including 3-site version)
  • Single-layer and double-layer PEPO reduced density matrix
  • spacetype method for InfinitePartitionFunction
  • Support for SU2Irrep symmetry in j1_j2_model

Changed

  • A unified interface for Trotter-based time evolution algorithms. The old su_iter, simpleupdate functions should be replaced by timestep, time_evolve respectively
  • Default fixed-point gradient algorithm changed to :EigSolver
  • BoundaryMPS methods now have their own custom transfer functions, avoiding a double conjugation and twist issues for fermions
  • physicalspace and related functions now correctly handle periodic indexing for infinite networks
  • Updated compatibility with TensorKit v0.15
  • Runic formatter

Removed

  • InfiniteWeightPEPS and mirror_antidiag
  • Support for integer space specifiers in state and environment constructors
  • Removed redefinition of tensorexpr
  • Support for dual physical spaces for non-bosonic symmetries

Fixed

  • Add unit normalization of the half and full infinite environments before the projector computation, fixing an issue with the gradient accuracy
  • Fix sporadic test timeouts when running variational optimization after simple update for the Heisenberg model by switching to a GMRES-based gradient solver
  • Rotation of iPEPO is now done correctly
  • Fix rotl90, rotr90 and rot180 for LocalOperator
  • Fix XXZ model convention
  • Fix add_physical_charge for fermionic operators
  • Fix maxiter behavior for fallback gradient linear solver
  • Fix gauge fixing in :fixed mode for non-uniform unit cells from full SVD

Performance

  • Avoid @autoopt for partition function calculations
  • Multithreaded scheduler now correctly taken into account for the reverse rules.