DuckHouse vs BigQuery.
Pay by the hour, or by the byte.
BigQuery’s on-demand pricing charges for the data each query scans: $6.25 per TiB after the first free TiB each month, with no machine to size. DuckHouse charges for the hours a database runs, however many queries it answers in that time.
BigQuery’s list prices for Iowa (us-central1), checked on November 1, 2026. Sources
Price list
Side by side.
Each service’s unit prices, in its own units. DuckHouse’s come from the code that bills it.
| Topic | DuckHouse | BigQuery |
|---|---|---|
| Compute | $0.0117 (512 MB) to $1.02 (16 GB) an hour | $6.25 per TiB scanned on demand, or $0.04 a slot-hour on Standard edition |
| Free each month | Nothing | 1 TiB of queries and 10 GiB of storage |
| Storage | $0.02 per GB-month in object storage (DuckLake), $0.225 for a DuckDB disk | $0.023 per GiB-month active, $0.016 long-term (logical bytes) |
| Smallest charge | The seconds a database runs | 10 MB per query, and per table a query reads |
Break-even
Hours against bytes.
What a month of each DuckHouse size costs, and how much on-demand scanning the same money buys in BigQuery, counting its free first TiB. Scan less than that and BigQuery costs less; scan more and DuckHouse does. Storage is left out: 100 GB costs about the same on both.
| DuckHouse size | 4 hours a working day | Same in BigQuery | Always on | Same in BigQuery |
|---|---|---|---|---|
| shared-cpu-1x (1 shared vCPU, 512 MB) | $1.03 | 1.2 TiB | $8.55 | 2.4 TiB |
| shared-cpu-4x (4 shared vCPUs, 2 GB) | $6.46 | 2.0 TiB | $53.55 | 9.6 TiB |
| shared-cpu-8x (8 shared vCPUs, 4 GB) | $13.69 | 3.2 TiB | $113.55 | 19.2 TiB |
| performance-4x (4 dedicated vCPUs, 8 GB) | $44.84 | 8.2 TiB | $372.00 | 60.5 TiB |
| performance-8x (8 dedicated vCPUs, 16 GB) | $89.69 | 15.4 TiB | $744.00 | 120.0 TiB |
Beyond price
Other differences.
| Topic | DuckHouse | BigQuery |
|---|---|---|
| Sizing | You pick a size, 512 MB to 16 GB, and can change it | Nothing to size on demand |
| What a query costs | Nothing extra; the hour costs the same busy or quiet | The bytes it scans, in the columns it reads, even with a LIMIT |
| SQL | DuckDB SQL; attach from any DuckDB client | GoogleSQL, through Google’s clients and drivers |
| Connectors | 15 sources synced on a schedule, at no extra charge | Not compared here |
Which to choose
Which one fits.
When BigQuery fits better
- Your use is light. 1 TiB of queries and 10 GiB of storage a month are free, and DuckHouse has no free tier.
- You don’t want to pick a size. On-demand BigQuery has no machine to choose or start.
- Your data is too big for one machine, or it already lives in Google Cloud.
When DuckHouse fits better
- Your dashboards and notebooks query all day. A DuckHouse database costs the same per hour however many bytes its queries read.
- You want a ceiling on the bill. The most a DuckHouse size can cost in a month is its hourly price for 730 hours, and stopping when idle brings it under that.
- You want DuckDB: its SQL, its file formats, and any DuckDB client attached over Quack.
- You want your sources synced for you. DuckHouse keeps 15 of them in step, including Postgres, MySQL, Stripe, Shopify, Square, QuickBooks, Google Analytics, Meta Ads, at no extra charge.
FAQ
About this comparison.
Where do these prices come from?
Does DuckHouse have a free tier?
Is a TiB in BigQuery the same as a TiB read by DuckHouse?
Sources
Where the numbers come from.
BigQuery’s prices are its published list prices for Iowa (us-central1), checked on November 1, 2026, and may have changed since. BigQuery is a trademark of its owner. DuckHouse isn’t affiliated with or endorsed by BigQuery.
Try it on your own data.
No contracts and no minimums. Pick a size, connect your sources, and stop the database whenever you like.