BrianQC 1.5 is coming
To be released in 2024 May
The new release have the following highlits:
-
Compatible with Q-Chem 6.1
-
Support for H100 is coming
-
full support for Pascal, Volta, Turing, Ampere, Ada Lovelace type of GPUs
-
external point charges are evaluated on GPU
-
electronic embedding of QM/MM is significantly faster
BrianQC 1.4
Released in December 2023
The new release have the following highlits:
-
Compatible with Psi4 1.4, 1.5 and 1.6 and Q-Chem 6.1
-
Added support of Ada Lovelace Nvidia GPUs: 40 series and L series (4090, L40 and similar)
-
upgraded CUDA libraries (12.2)
-
upgraded reaxFF support
-
several minor stability issues has been fixed
If you have any suggestions for new features or comments contact us at info@brianqc.com
BrianQC 1.3
Released in August 2022
The new release have the following highlits:
-
Removed support for NVIDIA Tesla K80 GPUs
-
Compatible with Psi4 1.4, 1.5 and 1.6 and Q-Chem 6.0
-
Added support for sp hybrid shells so STO and Pople type basis sets are faster
-
Improved small molecule performance
-
Completely reworked integrator system that lets us handle other type of integrals in the future (NMR)
-
Solved a few stability issues and a rarely appearing problem in the licensing system
If you have any suggestions for new features or comments contact us at info@brianqc.com
BrianQC 1.2.1 + PSI4 1.4 now on AWS cloud
Released in March 2022
Use BrianQC + PSI4 on AWS cloud without any costly HW or SW setup. Start a professionally installed and set up cloud instance with extremely good pricing.
Fall discount
12% discount will be applied to all license orders placed directly with BrianQC before November 30, 2021.
BrianQC 1.2.1
Released in September 2021
The new release have the following highlits:
-
Added support for NVIDIA Tesla A100 and NVIDIA GeForce RTX 30 series
-
Compatible with Psi4 1.4 and Q-Chem 5.4
-
Various bugfixes
-
Windows version is coming in the coming weeks!
All users having BrinaQC 1.2 on linux based system are asked to update to this version!
If you have any suggestions for new features or comments contact us at info@brianqc.com
BrianQC now supports the RTX 30 series
News in August 2021
We are proud to announce of the support of the Nvidia RTX 30 series. Below is a few results on an RTX 3070 Ti GPU (orange).
Its performance should be compared to the RTX 2070 (lime) and the GTX 1070 (dark blue).
BrianQC 1.2
News in June 2021
We are working on adding support for NVIDIA Tesla A100 GPUs.
Next in the line are the RTX 30 series.
BrianQC 1.2
Released in June 2021
The new release have the following highlits:
-
BrianQC now runs on Windows as well!
-
Added several new CPHF terms
-
Extended API
-
compatible with Psi4 1.4 and Q-Chem 5.4
-
SDK has now ReaxFF support
If you have any suggestions for new features or comments contact us at info@brianqc.com
Webinar with Q-Chem - already on YouTube!
News in August 2020
The new features of BrianQC 1.1 are in the focus. Watch video for discovering most recent news on BrianQC!
BrianQC 1.1
Released in June 2020
The new release have the following highlits:
-
Added support for more than 300 functionals:
-
Optimized DFT performance
-
CPHF calculations are now available
-
More easy to use API (application programming interface) and SDK (software development kit) capabilities
-
compatible with Q-Chem 5.3
If you have any suggestions for new features or comments contact us at info@brianqc.com
New DFT functionals
News in February 2020
The new release will contain more than 300 functionals including:
-
B3LYP
-
PBE, revPBE
-
TPSS, TPSSh,
-
SCAN
-
M06-L, M06-2X,
-
MN15,
-
B97, B97M,
-
ωB97X, ωB97M.
See full list here. Additionally BrianQC module from now on will support range separated, dispersion and nonlocal calculations as well.
If you have any suggestions for new features or comments contact us at info@brianqc.com
New feature directions
News in January 2020
Currently we are working on full CPHF support.
If you have any suggestions for new features or comments contact us at info@brianqc.com
BrianQC 1.0
Released in December 2019
The new release have the following highlights:
-
Stable DFT XC calculations for the following functional classes:
-
LDA
-
GGA
-
meta GGA
-
-
Improved DFT performance
-
One electron integral gradient
-
Geometry optimization
-
Molecule Mechanics support for single point and gradient
-
available force fields are: MMFF94 and UFF
-
-
DIIS algorithms for faster convergence
-
SDK features:
-
BrianQC functionalities can be accessed via an open API
-
several examples and samples are provided
-
-
Compatible with Q-Chem 5.2
BrianQC module 0.9 for Q-Chem 5.2
Released in May 2019
-
Added support for 1 electron integral calculations
-
Experimental support of DFT XC calculations for
-
LDA
-
GGA
-
mGGA
-
-
Support for Volta and Turing architecture
-
Optimized memory consumption
Extending our network of collaboration partners
News in March 2019
We are extending our network of collaboration partners throughout the world. Fill in your e-mail in the box below or click on Try BrianQC and download Q-Chem and BrianQC in a bundle for free to receive more discounts later.
BrianQC supports Tesla K40
Released in March 2019
We are dedicated to support our clients with their needs hence BrianQC now supports NVIDIA Tesla K40 upon customer request.
BrianQC now supports Volta architecture
Released in February 2019
The range of NVIDIA architectures supported by BrianQC has been expanded. In addition to GPUs powered by Kepler, Maxwell and Pascal, BrianQC now supports NVIDIA Tesla V100 GPU as well.
BrianQC featured on Q-Chem webinar series
News in December 2018
BrianQC module 0.8 for Q-Chem 5.1.2
Released in December 2018
BrianQC is a new Quantum Chemistry software, which is able to fully exploit the computational power of GPUs. Its innovative compiler technique will speed up your simulations and make your Q-Chem run on GPUs. BrianQC is the first GPU Quantum Chemistry software capable of calculating high angular momentum orbitals, and highly efficient for simulating large molecules and quantum systems.
-
Compatible with features of Q-Chem 5.1.2
-
Optimized for simulating large molecules
-
Tested up to 30,000 Cartesian Gaussian basis functions
-
Full support of s, p, d, f and g-type orbitals (h is coming soon)
-
Full support for NVIDIA GPU architectures (Kepler, Maxwell, Pascal)
-
Double precision accuracy
-
Multi GPU, cloud and supercomputer support
-
Runs on 64-bit Linux operation systems
-
Speeds up Q-Chem for every calculation that uses Coulomb or Exchange integrals over Gaussian basis functions or their first analytic derivative (including HF-SCF, DFT, SCF geom. opt, DFT geom. opt for most functionals, etc.)
See BrianQC at
BrianQC module 0.7 for Q-Chem 5.1.1
Released in August 2018
BrianQC is a new Quantum Chemistry software, which is able to fully exploit the computational power of GPUs. Its innovative compiler technique will speed up your simulations and make your Q-Chem run on GPUs. BrianQC is the first GPU Quantum Chemistry software capable of calculating high angular momentum orbitals, and highly efficient for simulating large molecules and quantum systems.
-
Compatible with features of Q-Chem 5.1.1
-
Optimized for simulating large molecules
-
Tested up to 20,000 Cartesian Gaussian basis functions
-
Full support of s, p, d, f and g-type orbitals (h is coming soon)
-
Full support for NVIDIA GPU architectures (Kepler, Maxwell, Pascal)
-
Double precision accuracy
-
Multi GPU, cloud and supercomputer support
-
Runs on 64-bit Linux operation systems
-
Speeds up Q-Chem for every calculation that uses Coulomb or Exchange integrals over Gaussian basis functions or their first analytic derivative (including HF-SCF, DFT, SCF geom. opt, DFT geom. opt for most functionals, etc.)
BrianQC module 0.5 for Q-Chem 5.0
Released in May 2018
BrianQC is a new Quantum Chemistry software, which is able to fully exploit the computational power of GPUs. Its innovative compiler technique will speed up your simulations and make your Q-Chem run on GPUs. BrianQC is the first GPU Quantum Chemistry software capable of calculating high angular momentum orbitals, and highly efficient for simulating large molecules and quantum systems.
-
Compatible with features of Q-Chem 5.0
-
Optimized for simulating large molecules
-
Tested up to 13,000 Cartesian Gaussian basis functions
-
Full support of s, p, d and f-type orbitals (g and h are coming soon)
-
Full support for NVIDIA GPU architectures (Kepler, Maxwell, Pascal)
-
Double precision accuracy
-
Multi GPU, cloud and supercomputer support
-
Runs on 64-bit Linux operation systems
-
Speeds up Q-Chem for every calculation that uses Coulomb or Exchange integrals over Gaussian basis functions or their first analytic derivative (including HF-SCF, DFT, SCF geom. opt, DFT geom. opt for most functionals, etc.)
May 2017
We are proud to be part of the Champions League of the European Union SME sector. We have been awarded the SME Instrument Phase 2 grant of European Union's Horizon 2020 under Grant Agreement No 767912.
March 2016
Our basic module is expected to be integrated and licenced by Q-Chem and goes to market in their new software release.
November 2015
QC Lab project scored over the threshold of the H2020 - SME Instrument Phase2 call obtaining Seal of Excellence certification delivered by the European Commission. The proposal also received recognition for the Hungarian Enterprise Europe Network. Since the last submission we were continuously working on the project, going further on the development and building business relationships, which are included now in the current proposal.
October 2015
New article published in Journal of Chemical Theory and Computation (Tornai, Gábor; Ladjánszki, István; Rák, Ádám; Kis, Gergely and Cserey, György “Calculation of quantum chemical two-electron integrals applying compiler technology on GPU" 2015).
New patent submitted for the Hungarian Intellectual Property Office (Simulation software consumer interface; P1500504) and previous national patent extended to Europe and USA.
May 2015
Thanks to our submitted publication to the Journal of Chemical Theory and Computation StreamNovation Ltd. started business negotiation with one of the biggest quantum chemistry simulator developers, Q-Chem Inc. Their objective is to integrate our QC Lab module into their widely used, market leader software. The collaboration has already started with regular discussions.
March 2015
Due to further development, integrator works for h type orbitals.
November 2014
Integrators work for d and f type orbitals, thus we can calculate on higher orbitals, therefore we can reach higher accuracy compared to the competitive solution, NWChem interface. The result was published in Chemical Physics Letters. (Rák, Ádám, and György Cserey. "The BRUSH algorithm for two-electron integrals on GPU." Chemical Physics Letters (2015).
October 2014
Business relation activities started with ChemAxon Ltd., Culevit Ltd. and Soft Flow Hungary Ltd.
September 2014
Our patent was evaluated with a very positive feedback by the European Intellectual Property Office.
March 2014
Integrators work for s and p type orbitals. Our new compiler technology works for single and double precision code generation. Based upon our numerical experiments, we concluded that our integration scheme leads to a speedup factor of 7-50x and 100-150x in case of double and single precision, respectively, meaning that a 3 weeks simulation run can be reduced for a few hours (depending on the molecule and the task). A demonstration of our approach took place together with several experts of quantum chemistry, who provided us feedback. They had the chance to test our program and calculate molecules of their interest. In their congruent opinion, our integration architecture could be highly advantageous compared to all the other alternative tools (regarding speed, accuracy, simulation run on high orbitals).
September 2013
Our approach has been compared to the GPL licensed quantum chemistry program NWChem and the non-free program MRCC. In each test case, our approach provided a sufficiently accurate result of chemically interesting quantities.
June 2013
Systems having only s and p type orbitals have been extensively tested. From our experiments, it has turned out that using our integrator architecture may lead up to three orders of magnitude speed-up.
May 2013
International patent has been filed under No PCT/HU2013/000051.