Supabase free project randomly slow? Check your Disk IO Budget
Ship ItChecked against the official source
Source checked 22 Oct 2026 · Published 23 Oct 2026 · Last reviewed 23 Oct 2026 · By Ctrl+Z Therapy (how we test)
The short answer
Supabase's docs say compute sizes up to 2XL, the free Nano included, can burst above their baseline performance for a short period of time every day. That is the Disk IO Budget: once it's consumed, your instance reverts back to its baseline performance, which on the free Nano compute is 5 MB/s and 250 IOPS (bursting to 261 MB/s and 11,800 IOPS). Supabase says response times on requests can then increase noticeably, CPU usage rises, and your instance may become unresponsive.
What to do
- In Supabase, open Observability, then Database Health.
- Look for the common causes Supabase lists: high memory use (swap), a low cache hit rate, and queries that take over a second.
- Optimize those queries first, or upgrade the compute.
- Ask your AI tool: "My Supabase project gets slow at random times. Find queries that take longer than a second and suggest indexes, before I pay for a bigger compute."
Good to know
- Not every slowdown comes from the disk budget, so check Database Health before you change anything.
- Supabase notes that Free plan compute is subject to change.
Source
Checked on 22 Oct 2026. Prices, limits and settings change, so check the source before you rely on them.