# Context size

AMALIA-9B is trained with a native context window of **32768 tokens** (`rope_theta 1e6`).

Included Modelfiles set `num_ctx`:
- `Modelfile.8k`  — 8192 (lightest RAM; good for chat)
- `Modelfile.32k` — 32768 (native maximum)

```bash
ollama create amalia-9b-32k -f Modelfile.32k
ollama run amalia-9b-32k
```

Larger contexts than 32K require RoPE scaling and will degrade quality — 32K is the honest native limit. KV-cache grows with context, so at 32K budget extra RAM on top of the weights.