Out of Memory
The fastest fix
Section titled “The fastest fix”Go down one model size.
That solves more Gemma 4 setup failures than changing runtimes, changing prompts, or re-downloading the same model again.
Typical signs
Section titled “Typical signs”- load hangs forever
- app crashes on load
- model initializes but never becomes usable
- the machine starts swapping heavily and everything becomes unusable
In order
Section titled “In order”- Try a smaller model.
- Try a lighter quantized variant if your runtime supports it.
- Close other memory-heavy apps.
- Use a runtime that makes hardware usage more obvious.
What not to do
Section titled “What not to do”Do not assume a runtime bug first. Most of the time the model simply does not fit cleanly on the chosen hardware.
Need the Gemma 4-specific version?
Section titled “Need the Gemma 4-specific version?”For the longer version covering long-context drag, KV cache growth, 16GB Mac tradeoffs, and which settings actually help, read Gemma 4 Out of Memory? Fix VRAM, RAM, and KV Cache Problems Fast.