Cloud-native OPM Flow simulation with AI that builds the deck for you. No Linux. No HPC. No procurement.
Your proxy model is in a spreadsheet. You know it is an approximation. Your client is making a multimillion Dollar decision on it.
Other solutions cost several thousand Dollar a year. Your company will not approve it. Your competitor already has it.
You downloaded OPM Flow. You spent three days on a Linux server configuring HPC. You never ran your first model.
You are using several solutions and none of them is really useful.
Full OPM Flow simulation. Not material balance. Not a type curve. A defensible dynamic reservoir model.
The same physics engine used in academic research and other commercial projects. Professional results at consultant pricing.
Browser-based. Sign up. Describe your reservoir. Run a simulation. Cloud compute included in your subscription.
All in a single place, from data loading, preparing to results viewing → in no time.
Scope → assumptions → run it → production dashboards → decision briefing
Full session · no edits · real OPM Flow run on EU cloud
Type any method, keyword or workflow, and tiles appear instantly.
Natural-language prompt → validated OPM Flow deck (reproducible in Runspec) → production dashboards → decision-ready PDF briefing. From scope to result in one session.
Form-based deck builder with guided section-by-section input. Grid · PVT · SCAL · Wells · Schedule — validated at every step.
Full keyword control with AI support.
Latin Hypercube Monte Carlo for uncertainty and sensitivity screening.
Create a simple anticline, crest at 3000m, 8° dip, N-S trending, 40×40×8 grid, Metric, 1× producer, OWC at 3200m and producing bhp controlled with 50 bar and 2× injectors injecting at initial pressure, for a 25 years forecast.
Tell SpectreAI the decision you're supporting — in plain English. No keyword syntax, no deck structure.
SpectreAI proposes every model assumption. Each value is visible, editable and confirmable. You stay in control. You get it documented in MD format.
One click. OPM Flow runs on managed cloud. FOPT, FWCT, GOR, pressure, 3D saturation maps — all in the browser.
SpectreAI delivers a structured engineering briefing — model rationale, confirmed assumptions, forecast, recommendation. Ready to present and ready to defend.
AI-generated models are reproducible in Runspec — no black box, fully transparent. Every keyword, assumption and grid visible and editable.
Not a simulation file — a decision document.
Black oil, gas condensate, and compositional models. Proxy studies, scenario planning, P10/P50/P90.
Saline aquifer injection, plume migration, capacity estimation. EU-mandate ready.
Doublet modelling, aquifer thermal energy storage, Enhanced Geothermal Systems. Not steam-dominated.
Hydrogen injection in depleted gas fields and saline aquifers. First cloud-native tool for H₂.
If you need a defensible answer by Friday, RUNSPEC is your tool. If you need full-field history matching with 200 wells — you still need other solutions. We will tell you that upfront.
Hetzner Online GmbH, Falkenstein & Nürnberg, Germany. EU jurisdiction. No US cloud providers for simulation data.
OPM Flow (GPL v3) — open-source, auditable, no vendor lock-in. Source: github.com/OPM
Anthropic Claude via EU Standard Contractual Clauses. Your simulation data is never used for AI model training.