Router weights fitted on 1,605 real IBM Heron jobs through 2 July 2026 — the full corpus, failures included. See the corpus →
GPU-first
Drug Discovery.
One-click discovery pipeline. Fold, generate and predict affinity run on GPU — nothing in the discovery chain has earned quantum hardware yet. The clinical chain already routes to it: trial-arm prioritisation runs QAOA on a real IBM Heron QPU and seals the job id with the result. Both are published, failures included.
01 Fold the target
Each panel draws the coordinates that day's run actually sealed — not a stand-in from a public database. Your browser re-downloads the model output, re-hashes it, and compares that hash to the one in the proof chain; the tick appears only if they match. Drag to rotate.
How it works
Seven sealed steps.
Two chains.
Open them.
Seven sealed steps across two chains: a three-step drug-discovery chain and a four-step clinical-evidence chain. Each step is saved so your team can review it before anyone touches a bench.
Pharma · molecule → IND
-
Fold the target
Build a 3D structure of the protein you care about with Boltz-2 (MIT), so later steps work from a real shape — not a guess.
-
Generate candidates
Propose small molecules aimed at that fold with GenMol (NVIDIA). You see what was suggested and how it sits in the pocket.
-
Predict affinity
Estimate how well the candidate may bind with Boltz-2 co-fold (MIT). Predicted — not a wet-lab measurement.
Rx · clinical evidence
-
Safety signal
Pull adverse-event signals from openFDA FAERS so safety context sits on the same trail as the rest of the evidence.
-
Evidence base
Assemble the clinical literature and prior findings your team will argue from — inspectable, not a black box.
-
Trial design
Size the study, pick the right patients, and lock the design assumptions before enrollment spend.
-
WORM clinical seal
Write the clinical record to a tamper-evident seal so compliance can open the trail later without taking our word for it.
Model layer · on by default
Sealed at the model.
Built for audit.
Every step is sealed at the model layer as it runs — signature and watermark on by default, written to a tamper-evident record for audit and traceability. Post-quantum cryptography. A third party can verify the trail without taking our word for it.
Device-aware routing
Time and money
on the right machine.
Wrong machine means queues, noise, and redos. Quantum when a GPU was enough wastes spend; GPU when you needed real hardware won’t survive diligence. EpochCore’s router weights were fitted on 1,605 real IBM Heron jobs through 2 July 2026 — a fit against recorded outcomes, not a live telemetry feed. Across those Heron devices the spread at p90 is 13×, which is the whole argument for choosing one. The dispatcher ranks live backends on queue depth and health today. No molecule-gen discovery step routes to a QPU; sealing that decision as a visible step, including when it says no, is the next build.
Two ways in
Discovery or trial.
Kill the bad molecule early.
Screen out non-starters before they burn years and budget.
Open Quantum Pharma →Design the trial. Carry the dossier.
Size the study, pick the right patients, keep the IND trail intact.
Open Quantum Rx →Who it’s for
R&D labs.
Built to be checked.
Built for biotech and discovery teams that need a run they can open and argue with. Not for clinical care or patient decisions.
- Discovery and computational chemistry groups evaluating new targets
- Biotech R&D leads who want a fold-to-binding trail before wet-lab spend
- Partner teams that need a shared, inspectable run — not a black box
What you can do today
Look first.
Or send a target.
Open the live run
Walk the saved fold, candidate, and binding steps in your browser. No signup.
Open the live runSend us a target
Tell us the protein you care about. We hand-run a discovery pass and return a link your team can open.
See a full discovery run
Open the live demo now, or email a target and we will run it for you.