Fabric F32 capacity
F32 suits a single department or a small analytics team running a handful of production models. Per-model memory is capped at 10 GB and every viewer still needs a Pro or PPU licence — the free-viewer threshold starts at F64.
- Pay-as-you-go
- $4,205/mo
- 1-yr reserved
- $2,501/mo
- Memory per model
- 10 GB
- Free viewers
- No — Pro needed
F32 limits at a glance
| Limit | Value |
|---|---|
| Capacity units (CU) | 32 |
| Equivalent Power BI SKU | EM3/A3 |
| Max memory per semantic model | 10 GB |
| Max memory per query | 5 GB |
| Model refresh parallelism | 20 |
| Max concurrent DirectQuery connections (per model) | 10 |
| Direct Lake rows per table | 300 million |
| Direct Lake Parquet files per table | 1,000 |
| Direct Lake max model size on OneLake | 40 GB |
| Viewers with free licence | Not allowed — Pro or PPU required |
| Copilot | Supported (not on trial capacity) |
What F32 costs
Fabric bills capacity units by the hour. At US East list prices, F32 costs $5.76 per hour, or about $4,204.80 for a full month. A 1-year reservation brings that to about $2,501.33 per month — roughly 40.5% less — in exchange for committing to the size for a year.
Below F64, the capacity cost is in addition to Pro or PPU licences for every viewer. F32 makes sense when you need Fabric workloads, larger models than Pro allows, or Copilot — not as a way to save on licences.
Prices are Microsoft list prices for US East, verified 2026-10-10, and vary by region and agreement.
Is F32 the right size?
Good fit for
- Departmental Import models up to roughly 5 GB each
- Direct Lake tables up to 300 million rows (40 GB max model size on OneLake)
- Mixing Power BI with lightweight Fabric workloads (Lakehouse, dataflows)
Watch out for
- Per-model memory of 10 GB — size the largest model, not the average
- Every viewer still needs Pro ($14/user/month) or PPU
- Refresh parallelism and DirectQuery connection limits during peak hours
The usual sizing mistake is choosing a SKU from the average model when the largest model and the morning refresh window decide whether a capacity throttles. The free Fabric capacity estimator checks both.
Compare neighbouring sizes
Fabric F32: common questions
- How much does Fabric F32 cost per month?
- At US East list prices, F32 costs about $4,204.80 per month pay-as-you-go ($5.76 per hour, 32 CUs × $0.18), or about $2,501.33 per month with a 1-year reservation — roughly 40.5% less. Prices vary by region and pay-as-you-go capacity can be paused when idle.
- Do report viewers need a Power BI Pro licence on F32?
- Yes. Below F64, every person who views Power BI content in a workspace on this capacity needs a Pro or Premium Per User licence. The free-viewer benefit starts at F64.
- How large can a semantic model be on F32?
- Each semantic model can use up to 10 GB of memory, including refresh overhead. Because a full Import refresh keeps the old copy available while the new one loads, a practical rule of thumb is to keep Import models below about 5 GB, or use incremental refresh. Each query is limited to 5 GB.
- What are the Direct Lake guardrails on F32?
- Up to 300 million rows and 1,000 Parquet files (and row groups) per table, with 40 GB max model size on OneLake. If a table exceeds them, Direct Lake on SQL endpoints falls back to DirectQuery (when fallback is enabled), while Direct Lake on OneLake fails to refresh until the table is optimised back under the limit.
- Can I use Copilot on F32?
- Yes. Copilot in Fabric and Power BI requires an F2 or larger capacity (or a P SKU) in a supported region; it is not available on trial capacities. Copilot usage consumes capacity units, so it competes with refreshes and queries on the same capacity.
Sources: Microsoft Learn — capacity limits, Direct Lake guardrails, Power BI pricing, Premium to Fabric migration, Copilot in Fabric. Verified 2026-10-10.