A developer reports high cache-miss rates on the Gemini API
Developer and creator Theo posted that Gemini is about 50% more expensive than it should be for his workload because caching isn't working well, citing a 27% cache miss rate. The post drew a lot of attention from other developers.
Why it matters
If you rely on prompt caching to keep costs down, a high miss rate quietly raises your bill.
Who it's for
Developers running high-volume Gemini API workloads.
What it means for you
- Log the cached token count that the Gemini API returns in each response's usage data.
- Work out your own hit rate over a day of traffic.
- Keep the cached prefix identical between calls (system prompt, tools, documents) and compare costs.
Applies to: Gemini API.
Note: reported by users on X; we haven't independently confirmed it works in the current version.
Source: @theo on X ↗
Written by AgentStackList from the original post. Checked Oct 17, 2026. Features, plans and availability can change, so confirm on the source before you rely on it.
gemini-apicachingcosts
More Gemini tricks
- Nano Banana prompt pack: SQL-style "planet in a hand" city dioramas · Oct 17
- A free year of Google AI Pro for college students (redeem by Dec 31) · Oct 17
- Turn a dense PDF into an interactive web app with Canvas · Oct 16
- Google AI Pro subscribers can now build with Argon in AI Studio · Oct 16
Agent Stack Weekly · free
Get the week's best AI tricks
Limit and pricing changes, new MCP servers, and setup fixes for Claude Code, Codex, Cursor, Copilot and Antigravity.
No spam. Unsubscribe anytime. Read past issues