SolvSRK-UQ
Release: Production (GA)
Single-solve uncertainty quantification for SolvSRK
What SolvSRK-UQ is
Simulation customers ask two questions: what is the state at time T? and how confident are you? Monte Carlo answers the second with thousands of re-runs. SolvSRK-UQ returns a terminal uncertainty envelope y(T) ± σ(T) from one integration pass — no MC farm at runtime.
It is the same libsolvsrk / solvsrk binary as SolvSRK.
Enable with emit_uq=True (Python) or emit_uq=1 (C) when your seat carries
the UQ entitlement.
Trial vs purchase
| Phase | UQ available? |
|---|---|
| 30-day SolvSRK trial | Yes — trial seats are issued with features: uq |
| Purchased SolvSRK (base) | No — integrator only |
| Purchased SolvSRK-UQ add-on | Yes — features: uq |
| Purchased SolvSRK-Edge | Yes — features: edge (implies UQ) |
SolvSRK-Edge includes UQ — no separate Edge UQ SKU. This page is the full guide for desktop/sim customers evaluating UQ on trial and buying the SolvSRK-UQ add-on after subscribe.
Validated on representative benchmark problems (libsolvsrk v2.4+ entitlement gate).
What it does
Simulation customers ask two questions: "what is the state at time T?" and "how confident are you?" Today the second question means Monte Carlo — thousands of re-runs, days of compute, expensive certification cycles.
SolvSRK-UQ adds a terminal uncertainty envelope y(T) ± σ(T) from a single integration pass.
No MC farm at runtime. Enable with one flag: emit_uq=True.
30-day SolvSRK trials include UQ. After purchase, SolvSRK and SolvSRK-UQ are separate SKUs — base seats need the UQ add-on (or Edge) for production envelopes.
When to use it
Evaluating envelopes during a SolvSRK trial; certification-heavy industries; real-time decision loops where MC is too slow; risk-quant teams on ODE models. SolvSRK-Edge includes UQ — this SKU is for desktop/sim customers adding UQ after they subscribe to base SolvSRK.
See Examples for runnable code.
Who it is for
- Teams evaluating SolvSRK who need terminal envelopes during the 30-day trial
- Certification-heavy industries needing defensible uncertainty bounds
- Real-time decision loops where Monte Carlo is too slow
- Risk-quant and regulatory-affairs teams pricing structured products on ODE models
What ships
- Same
libsolvsrk/solvsrkbinary as SolvSRK - Terminal envelope via
emit_uq/noise_sigma - Diagnostic trace via
uq_ek_handoff_trace() - Entitlement check:
solvsrk.license_has_uq()
Related products
Base integrator: SolvSRK. Included on Edge: SolvSRK-Edge.