One of Dorna's advantages is that you're not married to a single model: within the same workspace you switch among the available models, and your conversations and knowledge stay put.
But which model for which job? Our rule of thumb:
Light, frequent tasks → the faster model. Rewrites, short summaries, e-mail drafts, quick answers. These benefit from speed, the quality difference barely shows — and they consume less quota.
Heavy, high-stakes tasks → the stronger model. Contract analysis, multi-document synthesis, multi-step reasoning, copy going to an important client. Here quality is worth it.
Not sure? Start fast; if the answer lacks depth, repeat the same request with the stronger model. Conversations are saved, so comparing is easy.
Two things that matter more than model choice:
- Context beats model. An average model with the right knowledge base answers better than a strong model with no context — especially for questions that come back to your documents.
- Agents simplify the choice. When you hand a task to an agent, it drives the steps itself; you focus on defining the task well.
You'll see the available models right in the chat window after signing in — and if you don't have an account yet, the free plan is the place to start.