Create a free account to use the calculator

Members use every tool on AIDataLegends for free. It takes a name, an email and a password, and you come straight back here.

Not ready yet? The worked example below shows exactly what the tool produces.

Worked example

A Medium warehouse (4 credits/hour) at $3.00 per credit, running 10 hours a day on business days, with multi-cluster autoscaling between 1 and 3 clusters. The calculator's answer:

Total per month2,600.40 2,600.40–7,801.20 across the autoscaling range
WarehouseMedium 4 credits/hr per cluster
Per hour12.00 4 credits/hour total
Per year31,204.80 217 hours/month

What this means

  • This is compute credits only. The cloud services layer (query compilation, metadata, result caching) is free up to 10% of a day's compute credits and billed at the same per-credit rate beyond that — workload-dependent, so it isn't in this total. Storage is billed separately again, by the terabyte.
  • The range assumes the warehouse sits at one end the whole time. Real auto-scaling usually lands somewhere in between — use the range as bounds, not a midpoint estimate.
  • A warehouse resizing up or down takes effect on the next query, not retroactively, and billing jumps to the new size's rate immediately — resize for the workload you're about to run, not the one that just finished.

The breakdown

Credits/hour   = clusters running × credits/hour for Medium
               = 4
Cost/hour      = credits/hour × price per credit = 12.00
Cost/month     = cost/hour × hours/day × days/month = 2,600.40

Starting points, not an invoice: list prices vary by edition, cloud and contract, and this excludes the cloud services layer and storage. Use your own negotiated or published rates for an exact number.

How Snowflake warehouse billing works

Snowflake meters virtual warehouse compute in credits, consumed per second (60-second minimum) while the warehouse is running — not while it's suspended. Every size doubles the credit rate of the one below it:

SizeCredits/hour
X-Small1
Small2
Medium4
Large8
X-Large16
2X-Large32
3X-Large64
4X-Large128
5X-Large256
6X-Large512

A multi-cluster warehouse running N clusters at once consumes N times that rate — this calculator's cluster count models exactly that. The price per credit itself depends on your edition (Standard, Enterprise, Business Critical), cloud and whether you're on on-demand or capacity pricing.

What this doesn't include

  • The cloud services layer — query compilation, metadata operations and result caching — is free up to 10% of a day's compute credits and billed at the same rate beyond that. It's too workload-dependent to estimate here, so it isn't in this total.
  • Storage is billed separately, by the average terabyte per month, independent of compute entirely.
  • Serverless features (Snowpipe, Search Optimization, Automatic Clustering, materialized view maintenance) bill their own credits outside the warehouse you size here.

The levers that actually move this number

  • AUTO_SUSPEND. A warehouse left running idle bills every second it's up, whether or not a query is executing — a short auto-suspend window (Snowflake's default is a good starting point) is the single biggest lever most accounts haven't pulled.
  • Right-sizing. A warehouse resize takes effect on the next query at the new size's rate immediately — size for the query you're about to run, not the one that just finished.
  • Multi-cluster + auto-scale vs. a bigger size. Bursty concurrent-query load usually calls for more clusters of a smaller size (auto-scaling) rather than one bigger, always-on warehouse; steady heavy single queries call for the reverse.
  • Resource monitors cap credit spend per warehouse or account per period and can suspend warehouses automatically when a threshold is hit — useful as a backstop, not a sizing tool.

Limitations and assumptions

  • The autoscaling range assumes the warehouse sits at one bound (min or max clusters) the whole time; real auto-scaling workloads usually land somewhere between, so treat the range as bounds, not an estimate.
  • Cloud services credits, storage and serverless features are not included — see above.
  • Prices change by edition, cloud, region and contract, and this tool has no pricing data of its own — it is pure arithmetic on the numbers you provide.

Frequently asked questions

How much is one Snowflake credit?
There's no fixed dollar price — a credit's price depends on your edition (Standard, Enterprise, Business Critical), cloud provider, region and whether you're on on-demand or negotiated capacity pricing. Check your account's pricing page or billing dashboard for your actual rate.
Why does a Large warehouse cost exactly double a Medium one?
Snowflake warehouse sizes double the credit-consumption rate at every step (1, 2, 4, 8, 16, 32…), so compute cost for the same runtime doubles with every size increase regardless of edition or cloud.
What's the difference between warehouse size and cluster count?
Size sets the credit rate of one cluster. A multi-cluster warehouse can run several clusters of that same size at once to handle more concurrent queries — cluster count multiplies the per-cluster rate, it doesn't change it.
Is the cloud services layer ever worth worrying about?
Usually not — it's free up to 10% of a day's compute credits, and most workloads stay under that. Heavy use of many small, simple queries (lots of compilation relative to execution time) is the pattern most likely to push past the free allowance.
Why do I need an account?
The tools are free for members. Signing in lets us keep them free and fast, and protects them from automated abuse. Your input is used only to produce the answer and is not stored.

Keep going