Is Qemica really free?
Yes. The Free plan is the full workbench — every screen, every engine, no gated methods or molecule limits. In exchange, the Free plan includes your consent to de-identified calculation contributions (structure, settings, results, engine versions) used to improve calculation methodology and to train machine-learning models. Collection is not yet active — today no calculation data leaves your machine, and the app will tell you before it begins. Once active, contributions are severed from your identity at ingestion, never include your names, notes, or files, and are never sold. Premium excludes your results from contribution entirely. Details in the Privacy Policy.
What data leaves my computer?
Qemica is local-first: projects, molecules, and results are plain files on your machine. The launch-time update check is off unless you turn it on — you're asked on first run, and the box starts unchecked. Beyond that and a launch-time announcements fetch, the app touches the network only for things you explicitly trigger — sign-in, PubChem fetches, installing an optional capability, or running on hardware you point it at (IBM Quantum, your SSH cluster). Third-party credentials stay in your OS keychain and are sent only to that provider.
Can I move my work to a new computer?
Yes — Settings → Paths → Device migration exports everything (projects, results, settings) as a single zip file; import it on the other machine and your work appears there. It is a local file you move yourself: nothing is uploaded anywhere, and credentials stay in your OS keychain, so you re-enter those on the new device.
Do I need a quantum computer?
No. Every quantum method runs exactly on the local statevector simulator in the engine you install — no quantum hardware needed. If you do have IBM Quantum access, six methods can dispatch to real hardware with your own account and token — costs and quotas for that are between you and IBM.
Can I use my university cluster?
Yes — the Run screen can submit over SSH, including through SLURM or PBS queues with an editable submission-script template and per-job resources. Install the engine on the server with the same short uv setup as locally, and Qemica runs it there directly. Submission is detached: quit Qemica and it reattaches to the remote job on the next launch.
Why does Windows need WSL2?
The scientific Python stack Qemica drives (PySCF and friends) is Linux-native — there are no reliable native Windows builds. Qemica runs the engine inside WSL2 automatically and translates paths both ways; you install the engine environment once inside WSL following the setup steps.
Does it work offline?
Yes. Calculations run entirely on your machine, and a cached sign-in keeps working offline. You only need a connection for the one-time engine setup, PubChem fetches, optional-capability installs, and cloud targets.
Who owns my molecules and results? Can I publish them?
You do, and yes. Structures you create and results the software computes for you are yours under the EULA — publish freely. Results carry full provenance (method, basis, engine versions, RNG seed) so they are reproducible, and citing Qemica is appreciated.
Which methods and engines are included?
Classical electronic structure through PySCF (HF, DFT across Jacob's Ladder, MP2, CCSD, CCSD(T), FCI, CASSCF, TDDFT, and more), a dozen quantum eigensolvers plus vibrational and open-system dynamics on the local simulator, and the MACE machine-learned potential as a fast DFT surrogate. The full map is on the platform page.
How do updates work?
The app checks a version manifest at launch and shows a notice when a newer release exists — nothing installs silently. Engine updates are a one-line uv command shown in the app. See the changelog for what shipped.