Changelog - Revision history

Csheldon: /* 6.4.3 */

2026-07-21T08:51:38Z

6.4.3

← Older revision		Revision as of 08:51, 21 July 2026
Line 282:		Line 282:
	<div id="Changelog643" style="display:inline;"></div>		<div id="Changelog643" style="display:inline;"></div>

	==== FEATURE ====		=== FEATURE ===
	* Increased flexibility in the choice of exchange-correlation functionals: added the tags {{TAG\|XC}} and {{TAG\|XC_C}} to specify linear combinations of exchange-correlation functionals.		* Increased flexibility in the choice of exchange-correlation functionals: added the tags {{TAG\|XC}} and {{TAG\|XC_C}} to specify linear combinations of exchange-correlation functionals.
	* Additional MGGA functionals (v1-sregTM, v2-sregTM, v3-sregTM, and v2-sregTM-L) from Francisco, Cancio, and Trickey (https://doi.org/10.1063/5.0167868, https://doi.org/10.1063/5.0167873).		* Additional MGGA functionals (v1-sregTM, v2-sregTM, v3-sregTM, and v2-sregTM-L) from Francisco, Cancio, and Trickey (https://doi.org/10.1063/5.0167868, https://doi.org/10.1063/5.0167873).
Line 294:		Line 294:
	* {{TAG\|LWRITE_SPN}}=T writes the spin-matrix element to the .spn file for wannier90.		* {{TAG\|LWRITE_SPN}}=T writes the spin-matrix element to the .spn file for wannier90.

	==== IMPROVEMENT ====		=== IMPROVEMENT ===
	* Update [[makefile.include.nec_aurora]] template to work with the recent NEC compiler version (5.0.0+).		* Update [[makefile.include.nec_aurora]] template to work with the recent NEC compiler version (5.0.0+).
	* Add workarounds for Intel oneAPI LLVM compilers (ifx), and [[makefile.include]] files for these compilers.		* Add workarounds for Intel oneAPI LLVM compilers (ifx), and [[makefile.include]] files for these compilers.
Line 325:		Line 325:
	-->		-->

	==== BUGFIX ====		=== BUGFIX ===

	* Descriptor sparsification was not working in combination with the {{TAG\|ML_MODE}} tag: the tag {{TAG\|ML_LSPARSDES}} was automatically set to False.		* Descriptor sparsification was not working in combination with the {{TAG\|ML_MODE}} tag: the tag {{TAG\|ML_LSPARSDES}} was automatically set to False.

Csheldon at 08:51, 21 July 2026

2026-07-21T08:51:25Z

← Older revision		Revision as of 08:51, 21 July 2026
Line 1:		Line 1:
			== 6.6.1 ==
			<div id="Changelog661" style="display:inline;"></div>

			=== BUGFIX ===
			*Forces with [[METAGGA\|meta-GGA]] functionals were wrong in the spin-polarized case ({{TAG\|ISPIN}}=2). This bug was introduced in VASP 6.6.0 and is now fixed, cf. [[Known_issues#KnownIssue94\|known issue 94]].
			*[[Running_GRACE_force_fields_in_VASP\|GRACE force fields]] generated from <code>tensorpotential>=0.5.8</code> were incompatible with the VASPml interface, cf. [[Known_issues#KnownIssue94\|known issue 95]].
			*Born effective charges were wrong on GPU (OpenACC) with {{TAG\|LCALCEPS}}, though the dielectric tensor is unaffected; cf. [[Known_issues#KnownIssue93\|known issue 93]].
			*2d Coulomb truncations produced noisy potentials for [[KERNEL_TRUNCATION/LTRUNCATE\|Coulomb kernel truncation]] in 6.6.0 , cf. [[Known_issues#KnownIssue72\|known issue 72]].
			*OMP GPU offloading on AMD requires GPU pinning, cf. [[Known_issues#KnownIssue70\|known issue 70]].
			*Partial DOS were not computed correctly for {{TAG\|ISPIN\|2}} and {{TAG\|KPOINTS_OPT}}, cf. [[Known_issues#KnownIssue69\|known issue 69]].

	== 6.6.0 ==		== 6.6.0 ==
	<div id="Changelog660" style="display:inline;"></div>		<div id="Changelog660" style="display:inline;"></div>

Csheldon at 13:36, 20 July 2026

2026-07-20T13:36:08Z

← Older revision		Revision as of 13:36, 20 July 2026
Line 1:		Line 1:
	== 6.6.0 ==		== 6.6.0 ==
	<div id="Changelog660" style="display:inline;"></div>		<div id="Changelog660" style="display:inline;"></div>
	~~<!--~~We introduce NMR, restarting phonons, XAS, and cRPA in the release video ~~(ADD LINK)~~.~~-->~~		We introduce NMR, restarting phonons, XAS, and cRPA in the {{Video\|release66:chris:2026\|release video}}.

	=== FEATURE ===		=== FEATURE ===

Liebetreu at 11:01, 26 June 2026

2026-06-26T11:01:15Z

Liebetreu: /* IMPROVEMENT */

2026-04-13T14:01:20Z

IMPROVEMENT

← Older revision		Revision as of 14:01, 13 April 2026
Line 140:		Line 140:
	** Hermiticity of the BSE matrix is now enforced for {{TAG\|IBSE\|3}} to improve stability.		** Hermiticity of the BSE matrix is now enforced for {{TAG\|IBSE\|3}} to improve stability.

	* Improved the performance of the Ewald summation used in the truncated Coulomb kernel method for 2D materials by restricting the number of g-vectors that are used in the reciprocal space summation. This truncation of g-vectors is in keeping with the 3D summation, where every value less than <math>10^{~~−10~~}</math> is removed.		* Improved the performance of the Ewald summation used in the truncated Coulomb kernel method for 2D materials by restricting the number of g-vectors that are used in the reciprocal space summation. This truncation of g-vectors is in keeping with the 3D summation, where every value less than <math>10^{-10}</math> is removed.

	=== BUGFIX ===		=== BUGFIX ===

Singraber: /* FEATURE */

2026-03-31T22:17:34Z

FEATURE

← Older revision		Revision as of 22:17, 31 March 2026
Line 35:		Line 35:
	**Thermodynamic integration for particle insertion using MLFF and empirical potentials (ML_LEMPPOT, ML_EMPPOT_RCUT, ML_SRPOT_B0, ML_SRPOT_N0, ML_SRPOT_S0,ML_MOPOT_NM, ML_MOPOT_DM, ML_MOPOT_RM, ML_MOPOT_RKM, ML_MOPOT_IJM).		**Thermodynamic integration for particle insertion using MLFF and empirical potentials (ML_LEMPPOT, ML_EMPPOT_RCUT, ML_SRPOT_B0, ML_SRPOT_N0, ML_SRPOT_S0,ML_MOPOT_NM, ML_MOPOT_DM, ML_MOPOT_RM, ML_MOPOT_RKM, ML_MOPOT_IJM).
	**Add "delta" mode ({{TAG\|ML_MODE\|delta}}): always adding the prediction from a given {{FILE\|ML_FF}} to the ab initio calculation results.		**Add "delta" mode ({{TAG\|ML_MODE\|delta}}): always adding the prediction from a given {{FILE\|ML_FF}} to the ab initio calculation results.
	**Experimental support for GRACE force fields in prediction-only mode (requires VASPml, Tensorflow and cppflow).		**Experimental support for [[Running GRACE force fields in VASP\|GRACE force fields]] in prediction-only mode (requires [[Makefile.include#VASPml_(experimental)\|VASPml]], Tensorflow and cppflow).
	*[[:Category:HDF5_support \| HDF5]]		*[[:Category:HDF5_support \| HDF5]]
	**Write phonon frequencies and eigenvectors to {{FILE\|vaspout.h5}}.		**Write phonon frequencies and eigenvectors to {{FILE\|vaspout.h5}}.

Kaltakm: /* IMPROVEMENT */

2026-03-17T15:21:13Z

IMPROVEMENT

← Older revision		Revision as of 15:21, 17 March 2026
Line 59:		Line 59:
	**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).		**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).
	**Support for fast EXX mode in RPA and GW.		**Support for fast EXX mode in RPA and GW.
	**Double-counting corrections for RPA forces when using DFT+U		**Double-counting corrections for [[Known_issues#KnownIssue67 \| RPA forces when using DFT+U]].
	**Reduced memory usage in BSE calculations		**Reduced memory usage in BSE calculations
	**Improved support for single precision BSE with Lanczos algorithm.		**Improved support for single precision BSE with Lanczos algorithm.

Kaltakm: /* IMPROVEMENT */

2026-03-17T15:18:18Z

IMPROVEMENT

← Older revision		Revision as of 15:18, 17 March 2026
Line 59:		Line 59:
	**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).		**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).
	**Support for fast EXX mode in RPA and GW.		**Support for fast EXX mode in RPA and GW.
	**Double-counting corrections for RPA forces when using ~~PBE~~+U		**Double-counting corrections for RPA forces when using DFT+U
	**Reduced memory usage in BSE calculations		**Reduced memory usage in BSE calculations
	**Improved support for single precision BSE with Lanczos algorithm.		**Improved support for single precision BSE with Lanczos algorithm.

Wolloch: /* 6.6.0 */

2026-03-17T09:31:35Z

6.6.0

← Older revision		Revision as of 09:31, 17 March 2026
Line 64:		Line 64:
	*Change parameters in semilocal functionals ({{TAG\|XCm_Pn}}).		*Change parameters in semilocal functionals ({{TAG\|XCm_Pn}}).
	*Greatly improved stability of noncollinear GGA and metaGGA calculations.		*Greatly improved stability of noncollinear GGA and metaGGA calculations.
	*Allowed the Ewald cutoff parameter to be changed.		*Allowed the Ewald-cutoff parameter to be changed.
	*Implement {{TAG\|KSPACING}} for {{FILE\|KPOINTS_OPT}} file using KSPACING_OPT.		*Implement {{TAG\|KSPACING}} for {{FILE\|KPOINTS_OPT}} file using KSPACING_OPT.
	*New CMake build system as an alternative to the traditional makefile.include's. See [[Installing_VASP.6.X.X#Install_VASP_with_CMake\|Install VASP with CMake]] for instructions.		*New CMake build system as an alternative to the traditional makefile.include's. See [[Installing_VASP.6.X.X#Install_VASP_with_CMake\|Install VASP with CMake]] for instructions.

Singraber at 10:48, 16 March 2026

2026-03-16T10:48:27Z

← Older revision		Revision as of 10:48, 16 March 2026
Line 9:		Line 9:
	**[[LSCRPA \| spectral cRPA]].		**[[LSCRPA \| spectral cRPA]].
	**[[Constrained–random-phase–approximation_formalism#Off-centre_interactions \| multi-centre Coulomb matrix elements]].		**[[Constrained–random-phase–approximation_formalism#Off-centre_interactions \| multi-centre Coulomb matrix elements]].
	*Checkpointing of finite-difference calculations ({{TAG\|CHECKPOINT_FD}} when {{~~TAGO~~\|IBRION\|6}})		*Checkpointing of finite-difference calculations ({{TAG\|CHECKPOINT_FD}} when {{TAG\|IBRION\|6}})
	**[[Restarting_finite_differences_calculations#Restarting_a_finite_difference_calculation \| Restarting phonon calculations]].		**[[Restarting_finite_differences_calculations#Restarting_a_finite_difference_calculation \| Restarting phonon calculations]].
	**[[Restarting_finite_differences_calculations#Splitting_a_finite_difference_calculation \| Splitting phonon calculations]].		**[[Restarting_finite_differences_calculations#Splitting_a_finite_difference_calculation \| Splitting phonon calculations]].

← Older revision		Revision as of 10:48, 16 March 2026
Line 9:		Line 9:
	**[[LSCRPA \| spectral cRPA]].		**[[LSCRPA \| spectral cRPA]].
	**[[Constrained–random-phase–approximation_formalism#Off-centre_interactions \| multi-centre Coulomb matrix elements]].		**[[Constrained–random-phase–approximation_formalism#Off-centre_interactions \| multi-centre Coulomb matrix elements]].
	*Checkpointing of finite-difference calculations ({{TAG\|CHECKPOINT_FD}} when {{~~TAGO~~\|IBRION\|6}})		*Checkpointing of finite-difference calculations ({{TAG\|CHECKPOINT_FD}} when {{TAG\|IBRION\|6}})
	**[[Restarting_finite_differences_calculations#Restarting_a_finite_difference_calculation \| Restarting phonon calculations]].		**[[Restarting_finite_differences_calculations#Restarting_a_finite_difference_calculation \| Restarting phonon calculations]].
	**[[Restarting_finite_differences_calculations#Splitting_a_finite_difference_calculation \| Splitting phonon calculations]].		**[[Restarting_finite_differences_calculations#Splitting_a_finite_difference_calculation \| Splitting phonon calculations]].

← Older revision		Revision as of 11:01, 26 June 2026
Line 59:		Line 59:
	**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).		**Increased number of frequency points for the low-scaling GW (i.e., {{TAG\|NOMEGA\|24\|op=>}}).
	**Support for fast EXX mode in RPA and GW.		**Support for fast EXX mode in RPA and GW.
	**Double-counting corrections for [[Known_issues#KnownIssue67 \| RPA forces when using DFT+U]].		**Double-counting corrections for [[Known_issues#KnownIssue67 \| RPA forces when using DFT+U (issue 67)]].
	**Reduced memory usage in BSE calculations		**Reduced memory usage in BSE calculations
	**Improved support for single precision BSE with Lanczos algorithm.		**Improved support for single precision BSE with Lanczos algorithm.
Line 152:		Line 152:
	* Fixed inconsistent ML_FF input/output for non-scaLAPACK build.		* Fixed inconsistent ML_FF input/output for non-scaLAPACK build.

	* The code crashed for {{TAG\|LKPROJ\|True}}. This [[Known_issues#KnownIssue29 \| issue]] is fixed.		* The code crashed for {{TAG\|LKPROJ\|True}}. This [[Known_issues#KnownIssue29 \| issue 29]] is fixed.

	* SERTA is now correctly listed as "self-energy relaxation-time approximation" in {{FILE\|OUTCAR}}.		* SERTA is now correctly listed as "self-energy relaxation-time approximation" in {{FILE\|OUTCAR}}.

	* The <tt>wannier90.UNK</tt> files were not correctly written for non-collinear magnetic calculations. This [[Known_issues#KnownIssue31 \| issue]] is fixed.		* The <tt>wannier90.UNK</tt> files were not correctly written for non-collinear magnetic calculations. This [[Known_issues#KnownIssue31 \| issue 31]] is fixed.

	* If both {{TAG\|ISPIN\|2}} and {{TAG\|LNONCOLLINEAR\|True}} (or {{TAG\|LSORBIT\|True}}) were set in the INCAR, the code would stop with an error message (something about the number of entries for {{TAG\|MAGMOM}} being wrong). This [[Known_issues#KnownIssue26 \| issue]] is fixed: {{TAG\|LNONCOLLINEAR\|True}} will now take precedence over {{TAG\|ISPIN\|2}} (and the latter will default back to {{TAG\|ISPIN\|1}} internally).		* If both {{TAG\|ISPIN\|2}} and {{TAG\|LNONCOLLINEAR\|True}} (or {{TAG\|LSORBIT\|True}}) were set in the INCAR, the code would stop with an error message (something about the number of entries for {{TAG\|MAGMOM}} being wrong). This [[Known_issues#KnownIssue26 \| issue 26]] is fixed: {{TAG\|LNONCOLLINEAR\|True}} will now take precedence over {{TAG\|ISPIN\|2}} (and the latter will default back to {{TAG\|ISPIN\|1}} internally).

	* Fixed a problem with the final diagonalization in the occupied subspace before the computation of the forces: this bug sometimes caused errors in the forces when the wave functions were not tightly converged, for {{TAG\|ALGO\|A or D}}.		* Fixed a problem with the final diagonalization in the occupied subspace before the computation of the forces: this bug sometimes caused errors in the forces when the wave functions were not tightly converged, for {{TAG\|ALGO\|A or D}}.

	* BSE crashed on GPUs with <tt>-DCUSOLVERMP</tt> and <tt>-DCUBLASMP</tt> when {{TAG\|OMEGAMAX}} was set. This [[Known_issues#KnownIssue27 \| issue]] is fixed.		* BSE crashed on GPUs with <tt>-DCUSOLVERMP</tt> and <tt>-DCUBLASMP</tt> when {{TAG\|OMEGAMAX}} was set. This [[Known_issues#KnownIssue27 \| issue 27]] is fixed.

	* The computation of the dielectric function was not correctly implemented for finite-q in {{TAG\|IBSE\|1,3}}.		* The computation of the dielectric function was not correctly implemented for finite-q in {{TAG\|IBSE\|1,3}}.
Line 172:		Line 172:
	* The feature to compute the transport function using {{TAG\|ELPH_TRANSPORT_NEDOS_PLOT}} was broken. This has been fixed.		* The feature to compute the transport function using {{TAG\|ELPH_TRANSPORT_NEDOS_PLOT}} was broken. This has been fixed.

	* Electron-phonon code crashed for {{TAG\|ELPH_MODE\|renorm}}. This [[Known_issues#KnownIssue35 \| issue]] is fixed.		* Electron-phonon code crashed for {{TAG\|ELPH_MODE\|renorm}}. This [[Known_issues#KnownIssue35 \| issue 35]] is fixed.

	* VASP no longer crashes when running {{TAG\|ELPH_POT_GENERATE\|True}} without specifying the FFT grid via {{TAG\|ELPH_POT_FFT_MESH}}.		* VASP no longer crashes when running {{TAG\|ELPH_POT_GENERATE\|True}} without specifying the FFT grid via {{TAG\|ELPH_POT_FFT_MESH}}.

	* The {{TAG\|LATTICE_CONSTRAINTS}} were not applied correctly for {{TAG\|ISIF\|4, 5}}. This [[Known_issues#KnownIssue34 \| issue]] has been fixed.		* The {{TAG\|LATTICE_CONSTRAINTS}} were not applied correctly for {{TAG\|ISIF\|4, 5}}. This [[Known_issues#KnownIssue34 \| issue 34]] has been fixed.

	* For {{TAG\|ISIF\|4, 5, 6}}, the header of the {{FILE\|XDATCAR}} file was written twice. This has been fixed.		* For {{TAG\|ISIF\|4, 5, 6}}, the header of the {{FILE\|XDATCAR}} file was written twice. This has been fixed.

	* The truncated Coulomb kernel method for 2D systems only worked correctly if the cell was twice as large as the thickness of the slab. This requirement was imposed by the 2D Ewald summation, where minimum image convention was used in the aperiodic dimension. This requirement has now been removed by setting the distance between two atoms to not follow if the minimum image convention if it is 2D boundary conditions. This [[Known_issues#KnownIssue37 \| issue]] is fixed.		* The truncated Coulomb kernel method for 2D systems only worked correctly if the cell was twice as large as the thickness of the slab. This requirement was imposed by the 2D Ewald summation, where minimum image convention was used in the aperiodic dimension. This requirement has now been removed by setting the distance between two atoms to not follow if the minimum image convention if it is 2D boundary conditions. This [[Known_issues#KnownIssue37 \| issue 37]] is fixed.

	* There was a (small) memory leak in the use of HDF5. This [[Known_issues#KnownIssue37 \| issue]] has been fixed.		* There was a (small) memory leak in the use of HDF5. This [[Known_issues#KnownIssue37 \| issue 37]] has been fixed.

	* In case the HF exchange was activated via the {{TAG\|LTHOMAS}} tag, VASP wrongly kept {{TAG\|IMIX\|4}}. Now {{TAG\|IMIX\|1}} is used.		* In case the HF exchange was activated via the {{TAG\|LTHOMAS}} tag, VASP wrongly kept {{TAG\|IMIX\|4}}. Now {{TAG\|IMIX\|1}} is used.

	* The OpenACC version crashed when compiled with python plugin support (<tt>-DPLUGINS,</tt>). This [[Known_issues#KnownIssue30 \| issue]] is fixed and the python plugins are now fully supported in the OpenACC version as well.		* The OpenACC version crashed when compiled with python plugin support (<tt>-DPLUGINS,</tt>). This [[Known_issues#KnownIssue30 \| issue 30]] is fixed and the python plugins are now fully supported in the OpenACC version as well.

	* For {{TAG\|IBRION\|12}} or {{TAG\|PLUGINS/STRUCTURE\|True}} (i.e. using the python structure plugin) the updated structures were not written to {{FILE\|XDATCAR}}. This [[Known_issues#KnownIssue38 \| issue]] has been fixed.		* For {{TAG\|IBRION\|12}} or {{TAG\|PLUGINS/STRUCTURE\|True}} (i.e. using the python structure plugin) the updated structures were not written to {{FILE\|XDATCAR}}. This [[Known_issues#KnownIssue38 \| issue 38]] has been fixed.

	* The combination of {{TAG\|PLUGINS/STRUCTURE\|True}} and {{TAG\|ML_MODE\|run}} caused the code to crash. This [[Known_issues#KnownIssue36 \| issue]] is fixed.		* The combination of {{TAG\|PLUGINS/STRUCTURE\|True}} and {{TAG\|ML_MODE\|run}} caused the code to crash. This [[Known_issues#KnownIssue36 \| issue 36]] is fixed.

	* {{TAG\|ML_MODE\|train}} was broken when running on multiple nodes ''and'' using shared memory (<tt>-Duse_shmem</tt>). This [[Known_issues#KnownIssue28 \| issue]] is fixed.		* {{TAG\|ML_MODE\|train}} was broken when running on multiple nodes ''and'' using shared memory (<tt>-Duse_shmem</tt>). This [[Known_issues#KnownIssue28 \| issue 28]] is fixed.

	* Fixed broken ML_FF version check in VASPml; was unable to read ML_FFs from VASP 6.5.0 (earlier versions work).		* Fixed broken ML_FF version check in VASPml; was unable to read ML_FFs from VASP 6.5.0 (earlier versions work).
Line 198:		Line 198:
	* {{TAG\|ML_IERR}} was deprecated in favor of new tag name {{TAG\|ML_ESTBLOCK}}. This fixes "grepping" for <tt>ERR</tt> in {{FILE\|ML_LOGFILE}} (old tag still works).		* {{TAG\|ML_IERR}} was deprecated in favor of new tag name {{TAG\|ML_ESTBLOCK}}. This fixes "grepping" for <tt>ERR</tt> in {{FILE\|ML_LOGFILE}} (old tag still works).

	* The output of the total energy to the {{FILE\|OUTCAR}} was wrong when running {{TAG\|ML_MODE\|train}} with {{TAG\|ALGO\|A}}. This [[Known_issues#KnownIssue39 \| issue]] issue has been fixed.		* The output of the total energy to the {{FILE\|OUTCAR}} was wrong when running {{TAG\|ML_MODE\|train}} with {{TAG\|ALGO\|A}}. This [[Known_issues#KnownIssue39 \| issue 39]] issue has been fixed.

	* Fixed an incorrectly sized spline grid that sometimes led to wrong force predictions by VASPml.		* Fixed an incorrectly sized spline grid that sometimes led to wrong force predictions by VASPml.