← All troubleshooting topics

Performance & VRAM

When the GPU runs out of room or the UI feels heavy, there are a handful of dials that almost always fix it.

Ollama crashes with “out of memory” or the model fails to load

A pull works but loading or chatting fails with a CUDA / VRAM error.

1. Turn on Low VRAM mode in the Parameters sidebar (or globally in Settings → Features). This forces smaller batches and a leaner KV cache.
2. Lower num_ctx in the same panel — a smaller context window uses dramatically less VRAM.
3. Lower the GPU layer count to push more of the model onto CPU/RAM at the cost of speed.
4. Pick a smaller quantization (for example q4_K_M instead of q8_0) or a smaller parameter size.

The UI feels sluggish

Scrolling, typing, or window resizing is choppy.

• In Settings → Appearance, switch the theme from Aurora to Solid. Aurora’s animated gradient is heavy on weak GPUs.
• Close very long chats while testing — extremely long transcripts cost more to re-render on every token.

The first reply takes forever, even on a fast model

Sending the first prompt of the day takes 10+ seconds before streaming starts; subsequent replies are instant.

Ollama loads the model into VRAM on first use. To preload at launch, turn on Settings → General → Default model preload. The first chat will start fast, at the cost of pinning VRAM as soon as Loach opens.