Phase 3 · roadmap
open, citable fusion-ML benchmark suite — 3 tasks (CGYRO turbulence, MAST disruption, flux ROM) with real on-disk data, fixed reproducible splits, metrics, and KODEX baselines to beat. A community leaderboard starting point, not a private result; honest that it's small by mainstream-ML standards (pilot fusion-ML benchmarks).
KODEX KBENCH is a fast, calibrated surrogate for community-standard fusion-ML benchmarks — reproducing its result at inference speed, so it runs inside a real-time control loop or a design search where the full computation is far too slow to call.
One line, one contract — a prediction, its uncertainty, and whether the input is in-domain.
from kronos_ml import KBENCH
model = KBENCH() # loads the trained surrogate
y, sigma, in_domain = model.predict(x) # y = benchmark prediction
if not in_domain: # out of its trusted region
fall_back_to_full_physics() # KBENCH abstains, never extrapolatesEvery KODEX code wraps the shared spine — KHALO for calibrated uncertainty and KGATE for the out-of-domain gate — so it reports how confident it is and abstains rather than extrapolate. Benchmarks are computed on held-out data with a fixed seed; pre-registered misses are kept, not hidden.
Part of the open kronos-ml package (Apache-2.0). Open-access deposits with citable DOIs are listed below.
Open-access deposits, each with a citable DOI: