Cline
Cline supports OpenAI-compatible providers.
- Create a key in the console, for example “Cline”.
- In Cline open the settings (⚙️ icon) and choose API Provider: OpenAI Compatible.
- Fill in:
- Base URL:
https://api.beatriceai.it/v1 - API Key: your
bea_sk_… - Model ID:
beatrice-flash
- Base URL:
- In Model Configuration set:
- Context Window:
262144 - Max Output Tokens:
32768, or less if you want shorter answers
- Context Window:
Tips
- Cline sends long contexts at every step: thanks to the cache, the repeated part doesn’t count towards input limits.
- The model reasons before acting: leave enough output room, or reasoning can use it all.
- If you get 429s on a plan with one request at a time, avoid running several Cline tasks in parallel.