The future,priced.
Send your numbers. Get back tested answers: value at risk, tail risk, reserves, simulations and climate data.
See it work
One request. One clear answer.
These are real responses from the server. Pick an example.
curl api.moluscca.com/v1/risk/var \
-H "Authorization: Bearer $KEY" \
-d '{
"returns": [0.0041, -0.0123, 0.0067, ...],
"value": 250000,
"confidence": 0.99
}'curl https://api.moluscca.com/v1/risk/var \
-H "Authorization: Bearer $MOLUSCCA_KEY" \
-H "Content-Type: application/json" \
-d '{
"returns": [0.0041, -0.0123, 0.0067, 0.0012, -0.0215, 0.0089, 0.0034, -0.0051, 0.0102, -0.0008, 0.0046, -0.0177],
"value": 250000,
"confidence": 0.99
}'{
"object": "risk_var",
"observations": 12,
"confidence": 0.99,
"historical": { "var": 5270.5, "tvar": 5375, "tail_observations": 1 },
"normal": { "var": 6467.567793, "tvar": 7354.128353 }
}curl api.moluscca.com/v1/simulate/montecarlo \
-H "Authorization: Bearer $KEY" \
-d '{
"model": "gbm",
"params": { "s0": 100, "mu": 0.06, "sigma": 0.2 },
"paths": 5000,
"seed": 7
}'curl https://api.moluscca.com/v1/simulate/montecarlo \
-H "Authorization: Bearer $MOLUSCCA_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "gbm",
"params": { "s0": 100, "mu": 0.06, "sigma": 0.2 },
"horizon": 1,
"steps": 4,
"paths": 5000,
"seed": 7,
"sample_paths": 0
}'{
"object": "monte_carlo",
"model": "gbm",
"paths": 5000,
"terminal": { "mean": 105.7183716, "p5": 74.56899992, "p50": 103.1562102, "p95": 144.9727786 },
"theory": { "mean": 106.1836547 }
}curl "api.moluscca.com/v1/climate\ ?lat=13.08&lon=80.27\ &from=2025-04&to=2025-06\ &vars=irradiance,rainfall" \ -H "Authorization: Bearer $KEY"
curl "https://api.moluscca.com/v1/climate?lat=13.08&lon=80.27&from=2025-04&to=2025-06&vars=irradiance%2Crainfall" \ -H "Authorization: Bearer $MOLUSCCA_KEY"
{
"object": "climate_series",
"sample": true,
"location": { "name": "Chennai", "distance_km": 0 },
"series": [
{ "month": "2025-04", "irradiance": 6.79, "rainfall": 10 },
{ "month": "2025-05", "irradiance": 6, "rainfall": 34.3 },
{ "month": "2025-06", "irradiance": 5.15, "rainfall": 56.6 }
]
}What it is
Send your numbers, get answers you can trust.
Risk maths is easy to get almost right and hard to get exactly right. We did the hard part once, tested it against known answers, and put it behind a simple web address.
Plain HTTP and JSON
Works from Python, JavaScript, Excel or a no-code tool. No library to install, nothing to keep updated.
Your data is not kept
We compute the answer in memory and drop your numbers straight after. Only a usage count is logged.
Checked against known answers
Each method is tested against hand calculations or published results before it goes live.
Twelve methods today. One habit: nothing ships until it matches a known answer.
Maths
The calculations risk teams actually use.
Twelve endpoints live today, grouped the way you think about the problem. More are on the way, and we only list them once they pass their tests.
Distributions
Find the curve that best describes your data, then read off any percentile you need.
Value at risk and tails
How bad a bad day, a bad year or a 1-in-200 event could get, measured three ways.
Simulation
Thousands of possible futures for a price, a rate or a loan book, repeatable to the last digit.
Reserving and insurance
How much to hold for claims, and how a whole year of losses could add up.
Statistics
The basics done properly: summaries, regressions and models for counts and yes-or-no outcomes.
SaaS metrics
Revenue, churn, retention and payback from a simple list of customers and what they pay.
Data Early access
Climate risk data for India.
Monthly sunshine, heat, humidity and rainfall for any point, from one request. It helps insurers price weather risk, lenders understand farm and property exposure, and solar teams estimate output.
{
"location": { "id": "IN-CHN", "name": "Chennai", "lat": 13.08, "lon": 80.27, "distance_km": 0 },
"sample": true,
"series": [
{ "month": "2025-04", "irradiance": 6.79, "rainfall": 10 },
{ "month": "2025-05", "irradiance": 6, "rainfall": 34.3 },
{ "month": "2025-06", "irradiance": 5.15, "rainfall": 56.6 }
]
}How it works
Three steps, no setup.
Get a key
Sign up and copy your key. Free comes with 1,000 credits a month, no card needed.
Send your numbers
POST a few lines of JSON from any language: returns, claims, a triangle or a list of customers.
Get answers back
Clean JSON with the result, the method used and any warnings about your data. Ready for a dashboard or a report.
Pricing
Start free. Pay as you grow.
Every plan has every maths endpoint. You pay for how much you use, not for features you have to unlock.
Free
For trying it out and small experiments.
- 1,000 credits a month
- Every maths endpoint
- Community support
Starter
For side projects and early products.
- 25,000 credits a month
- Every maths endpoint
- Email support
Growth
Most teamsFor teams running it in production.
- 150,000 credits a month
- Climate data included
- Email support within one business day
Scale
For products that lean on it every day.
- 750,000 credits a month
- Climate data included
- Priority support
- 99.5% uptime SLA
Enterprise
For bulk data and models built for you.
- Credits sized to your volume
- Bulk climate data
- Custom models
- A named contact
What is a credit?
Most requests cost 1 credit. Big simulations cost more, and we show the cost before you run them. A Monte Carlo run with more than 10,000 paths costs 10.
If you go over
Paid plans keep working. Extra credits cost $1.50 (Starter), $1 (Growth) or $0.70 (Scale) per 1,000. On Free, requests pause until next month.
Billing
Pay monthly, or yearly and get 2 months free. Cancel any time. Payments run through Dodo Payments. INR prices are plus 18% GST.
INR overage rates will be confirmed before launch. Errors never cost credits.
Questions
Good to know.
Is my data stored?
No. Your numbers are processed in memory to compute the answer and then dropped. We only log the key, endpoint, time, speed and size of each request, for billing and security.
Where does the maths run?
On our servers. You never download the models, so they stay up to date and you do not have to maintain them.
How do I know the numbers are right?
Every endpoint is tested against hand calculations or published results, for example the chain ladder against the RAA triangle in Mack (1993). The tests run on every change.
What happens if I go over my credits?
On paid plans you keep going and pay a small amount per extra 1,000 credits. On Free, requests pause until the month resets.
Can I cancel any time?
Yes. Cancel from your account and the plan stops at the end of the period you paid for.
Is the climate data real?
Climate data is in early access. Today the endpoint serves a clearly labelled sample so you can build against the shape of the API. Every response says sample: true until the real dataset is live.
Do I need to be a quant to use it?
No. If you can send JSON you can use it. The docs explain what each number means in plain words.
Ready when you are.
A free key takes a minute. Your first answer takes one request.