Actuarial valuation & ALM platform for insurance carriers
Valuations, reserves and full ALM projections in seconds to minutes on a single workstation. Nested stochastic VM-21 in seconds, not days. Your data formats, your network, your operating model — nothing to rip out.
Why the name InstaVal?
Every one of these is measured or shipping today. The rest of this page is the evidence.
The opportunity
Most carriers run an actuarial engine plus a warehouse, ETL, BI and reconciliation layer built around batch runtimes. InstaVal drops in beside all of it — nothing has to change on day one.
The machinery around the engine can shrink at your pace, as far as you want — the savings are real, but they are your call, on your timeline.
Coverage
Assets and liabilities projected together on one engine, with the statutory frameworks and strategy solvers built in.
Strategy solvers search over full projections: Defeasance delivers the market-consistent measurement that enables IFRS 17 Phase 2; Capital Funding solves funding IRR and required capital at a hurdle rate.
Life & health
Term, whole life, UL / IUL / VUL, DI and LTC all value the same way: define the states a contract can be in and the transition probabilities between them, and InstaVal projects the full state distribution every period. A new product is a new set of states and transitions, not a new engine.
Full state distributions, not point-estimate decrements — death, lapse, disability and recovery compete correctly every period.
The audit workbook shows every state and transition for any policy, so an actuary can verify the projection by hand.
InstaStudy · New in v2.10
Experience studies run against the assumptions you actually value with, and the measured basis goes straight back into the model. Actual-to-expected means actual versus what you assume, not versus some other table.
Today an experience study is a spreadsheet or a one-off stats job: disconnected from the valuation basis, redone by hand each cycle, hard to audit. InstaStudy runs it from the admin extract, against the same store, annual or monthly, on policy-year or claim-clock durations. Five million policies and 166 million exposure cells run end to end in under a minute on a desktop workstation (internal measurement, synthetic data, October 2026).
Mortality, lapse, surrender, DI and LTC claim incidence by cause and care level, recovery, and disabled-life / on-claim mortality.
Independent-decrement rates on actuarial exposure, per decrement — the classic method, with millions of policies as the routine case.
DI and LTC on the claim clock — duration since claim commencement, monthly or annual — aligned with InstaVal's claim tables so import and publish round-trip exactly.
States and decrements come from one dictionary validated by both the study engine and the valuation, so the two vocabularies cannot drift.
Schema, data quality, rate coverage and exposure basis are checked before a number exists; holes are reported on publish or filled from the expected basis, never with zero.
Every study is a set of plain columnar tables any BI tool can read — exposure detail included — with a provenance row recording the run.
Shipping now in InstaVal v2.10, inside the binary: same download, same license, same credits. No second product to install. The bundle includes a sample study block, so the whole workflow runs in about a minute.
InstaReport · New in v2.23
InstaReport turns two period-end valuations into the period's story: the reserve roll-forward, the drivers of earnings, the transitions, and one row per contract, on one basis.
InstaReport values the opening book as a projection on a best-estimate basis, values the closing in-force book, and attributes everything in between. The valuations it runs are your period-end valuations: priced exactly as if you had run them on their own, with the results kept, plus a report line per contract for the attribution. Nothing is charged twice. Already ran your valuation this quarter? Next time, run the report as the valuation.
The unexplained line is zero on every in-force contract. Deaths, lapses, new issues and data breaks come straight from the end-of-period file.
The same start- and end-of-period inventories a valuation takes, in CSV, Parquet or Excel, over any whole number of months.
Assumption sets and discounts per date, a best-estimate basis of its own for the expected side, optional actual cash flows and a block investment table, and the choice of which waterfall steps to run.
Term, whole life and universal-life products run together as one report.
The transitions, and every contract's movement, paged and filtered on screen, behind every total in the report.
Each table as CSV, Parquet or Excel, one workbook with all of it, and the runs' own per-contract results, which stay with the report.
Shipping now in InstaVal v2.23, inside the binary: same download, same license, same credits. The bundle includes a sample period (a term book a year apart, with deaths, lapses, data breaks and new issues), so you can run a whole report before loading your own.
Adoption
If your admin system can export a CSV, you're ready to run — you're likely loading your first portfolio within hours, not weeks. Conversion isn't a cliff, it's a gradual ramp: start with one product type, validate against your current system, expand.
Governance
The audit trail isn't a defended artifact. It's a reproducible computation — keep what's convenient, regenerate what's questioned.
Audit & compliance
Every run is identified, versioned and reproducible. Same inputs, same outputs — guaranteed.
Any single contract: per-period cash flows, state distributions, 3-basis reserves. In milliseconds.
Multi-sheet Excel export: policy features, state model, transition matrices, scheduled amounts, projection results, discount rates.
An auditor asks how you got a number — you rerun the exact calculation in front of them.
Sequences
A sequence is a versioned bundle of inputs — portfolio, assumptions, scenarios, parameters. Vary any input across a sequence and the platform runs the sweep: sensitivity analysis on anything, not just the shocks your vendor pre-built.
Shock any assumption, rate or parameter — no special project, no custom code. Define the sweep, click run.
Every leg has a sequence ID and versioned inputs. Reproduce any cell of the sensitivity grid, on demand.
Each leg runs in seconds to minutes, so a 20-leg sensitivity grid is an hour at most — not a weekend of batch scheduling. When a leg is cheap, you stop rationing questions.
Regulatory
US statutory modules are validated against independent reference calculations.
The numbers
Legacy engines run overnight — or over a weekend — and you review the results tomorrow. InstaVal runs in the same sitting: run, look, change an assumption, run again.
Scale — September 2026
Full ALM book, closed / production mode, 14 product classes (79% liabilities) — on an owned workstation with a ~$1K NVIDIA card, not a datacenter.
Cold vs. steady: the first scenario after boot is ~9 min (526 s, measured), then ~2 min each — warm-up is paid once per server start. From a fresh install, the first bond valuation takes 1.4 s on a Mac Studio M5 Max (v2.22.2).
No GPU? The valuation / reserve API scales with cores — 27× at 32 cores, 46× at 64 on a rented x86 server, CPU only.
Deployment
Legacy: 3–5 days on a dedicated cluster, quarterly at best. InstaVal: 14 seconds on one workstation, on demand, from the VM-21 tab or the API — 10K contracts × 1,000 outer scenarios × 10 valuation dates × 50 inner scenarios, measured on a Mac Studio M5 Max (7.1 s on an RTX 4070 Ti), no model-point compression.
The Mac Studio M5 Max is faster per record than an RTX 4070 Ti workstation on every product we measured at book scale: 1.4–2.9×.
What is timed: a whole book at production scale — 1M-record valuation books and 500K-record projection books, every reserve basis on, warm. Microseconds per record; lower is faster. Measured October 2026. Same engine, same outputs to the cent. * Pensions and structured settlements run on CPU on both machines. Nested stochastic VM-21 ASPA (10K contracts × 1,000 outer × 10 dates × 50 inner, wall-clock, no model-point compression): 14 s on the Mac Studio M5 Max, 7.1 s on the RTX 4070 Ti. Pensions, 100K participants, CPU: valued in 0.69 s on one basis, 1.07 s on two; three bases plus a 30-year projection in 2.7 s.
| Product | Run | Apple · Mac Studio M5 Max | NVIDIA · RTX 4070 Ti workstation | Faster |
|---|---|---|---|---|
| Term life | Valuation | 8.3 µs | 16.1 µs | M5 Max 1.9× |
| Term life | Projection | 27.8 µs | 52.8 µs | M5 Max 1.9× |
| Disability income | Valuation | 19.8 µs | 35.6 µs | M5 Max 1.8× |
| Disability income | Projection | 125.4 µs | 200.2 µs | M5 Max 1.6× |
| Long-term care | Valuation | 16.1 µs | 32.0 µs | M5 Max 2.0× |
| Long-term care | Projection | 97.1 µs | 182.8 µs | M5 Max 1.9× |
| Annuities | Valuation | 12.6 µs | 24.3 µs | M5 Max 1.9× |
| Annuities | Projection | 19.3 µs | 38.9 µs | M5 Max 2.0× |
| VM-21 CTE70 | 51 scenarios | 4.0 µs | 8.1 µs | M5 Max 2.1× |
| Pensions (DB) * | Valuation | 18.6 µs | 40.9 µs | M5 Max 2.2× |
| Pensions (DB) * | Projection | 18.5 µs | 26.1 µs | M5 Max 1.4× |
| Structured settlements * | Valuation | 24.8 µs | 48.3 µs | M5 Max 1.9× |
| Structured settlements * | Projection | 23.6 µs | 51.1 µs | M5 Max 2.2× |
| Bonds | Valuation | 5.7 µs | 15.3 µs | M5 Max 2.7× |
| Mortgages | Valuation | 6.7 µs | 19.6 µs | M5 Max 2.9× |
Why this is hard to copy
Even if your current vendor got 100× faster tomorrow, every carrier's warehouse, ETL, BI and reconciliation stack would still exist — because those systems are built around the assumption that results are artifacts to be stored.
See it yourself
In the hosted demo, run your first model in the first five minutes.
Next steps
Conversion cost is the usual reason to wait. So the person who built InstaVal works alongside your team — from the first extract to the first reconciled run, with the data mapping documented and the reconciliation on file for your model governance and your auditors.
One product block from your book. We map your admin extracts and assumption tables to InstaVal's inputs with you.
Your portfolio, your workloads, your hardware. Measure the wall-clock time and reconcile to your current numbers.
On your own workstation or server, alongside your current system, until your team signs off.
The pilot is free — the license and our help. Converting the rest of the book is a separate engagement, scoped and priced with you.
How licensing works
InstaVal runs on your own hardware — Mac (Apple silicon) or Linux (NVIDIA GPU) — as a single self-contained bundle, air-gap capable. Usage is metered in credits: every valuation or projection consumes credits in proportion to what it computes (policies × scenarios × product complexity), debited before the run and refunded automatically if the engine errors. Scenarios bill sublinearly, so a 10,000-scenario run costs far less than a hundred times a 100-scenario one. Nothing leaves your machine except an optional usage record.
Every tier runs the same meter, and a credit is one dollar — the number on your dashboard is the number on your invoice. Standard licenses are prepaid: buy credits in the portal whenever you need more, no subscription, no per-seat license. Enterprise licenses are an annual contract with uncapped usage — run as much as you like, with no balance to watch — billed at the frequency you prefer; the meter keeps reporting consumption so the license can be sized to your book. Terms are scoped in a conversation, not on a price list.
A hosted demo instance for prospects, then a no-charge pilot on your own hardware with an evaluation license.
Metered per run, fractional credits, no ceiling and no minimum term on Standard.
Air-gapped installs replenish by pasting a signed token. Nothing phones home unless you let it.
A vendor-hosted cloud tier is on the roadmap and will be built when customer demand warrants it. Today, the same binary runs on a GPU VM you rent in your own cloud account — on-prem under the license, with none of the idle cost.
Why us
InstaVal was built by someone who spent years on the wrong side of this problem — helping carriers navigate their data frustrations, watching actuarial teams wait hours for results they needed in minutes. Every design decision, from the engine to the web UI, was made by someone who knows what it feels like to be the person waiting for the batch job to finish.
A 30-minute conversation: we walk you through the platform, run a live valuation against a portfolio shape that matches yours, and agree what a free proof of concept on your actual book would look like.