Choosing a model
Which model backs the agent, which plan reaches it, and how to match it to the task.
The agent is backed by a model you choose, and you can change it any time from the picker under the chat input. The model shapes how the agent reasons, how well it handles hard problems, and what the work costs. Picking well is mostly about matching the model to the task.
The models
| Model | Good for |
|---|---|
| Auto | The fast default. Free and unmetered on every paid plan. Routine work like reformatting, simple transforms, "what's the shape of this," and straightforward plots. |
| DeepSeek V4 Pro | A capable open model with adjustable reasoning. A solid step up from Auto for everyday analysis. |
| Kimi K2.6 | Another strong open model. |
| Kimi K2.7 Code | The newest Kimi, tuned for code, with a large context window. |
| Sonnet 5 | Claude Sonnet, a frontier model that handles real analysis and modeling well. |
| Opus 4.8 | Claude Opus, the most capable. For genuinely hard problems: subtle debugging, designing an approach, multi-step work where a wrong turn is costly. |
| Opus 4.7 | The previous Opus, also available. |
| GPT-5.5 | OpenAI's frontier model, if you prefer it or want a second opinion on a hard problem. |
Kimi K2.7 Code has a Fast option, and it is not its own row. Select K2.7 Code and a ⚡ Fast toggle appears on it: the same model served at roughly twice the speed, for about twice the token price. Turn it on when you're iterating and the waiting is what hurts.
Which plan reaches which model
This is the mapping, and it's the one the rest of the docs point at.
| Plan | Models you can pick |
|---|---|
| Free | Auto, and only Auto. The picker is locked to it. |
| Plus | Auto, DeepSeek V4 Pro, Kimi K2.6, and Kimi K2.7 Code with its Fast toggle. |
| Pro | Everything on Plus, plus the frontier models: Sonnet 5, Opus 4.8, Opus 4.7, GPT-5.5. |
| Max | The same models as Pro, with a larger monthly allowance. |
Two routes go around that table entirely, and both work on every plan, including Free:
- Bring your own key — connect an Anthropic or OpenAI key and Sonnet 5, Opus 4.8, or GPT-5.6 unlock, billed to your provider account. GPT-5.6 is only available this way.
- Local runner — pair your own machine and Clusy drives the
claudeorcodexCLI you already have, on the subscription you already pay for.
Neither one draws down your Clusy allowance. See Plans for what else each tier includes.
How to choose
A rough rule that holds up:
| The task is… | Reach for |
|---|---|
| Mechanical, like a reformat, a rename, or a simple plot | Auto |
| Normal analysis and modeling | DeepSeek V4 Pro or Kimi |
| Hard, like a tricky bug, approach design, or a high-stakes step | Sonnet, Opus, or GPT-5.5 |
You're not locked in. Start a session on Auto, switch up when you hit something that needs more, and switch back down after. Because the choice is per-message, you spend the expensive model only where it earns it.
The model is one dial, and reasoning effort is the other. A mid-tier model on high effort can beat a top model on low effort for some tasks, and cost less while doing it. Both are worth understanding.
Defaults and custom instructions
Set a preferred model and effort in Settings → Models so new sessions start where you like, and per-message overrides still apply on top. The same screen has custom instructions: standing guidance sent with every request, such as your conventions, preferred libraries, and plot style. You get up to 2,000 characters, set once instead of repeating yourself.