Choose a model
How to pick between Dropstone Fast, Pro, and Heavy, where the selector lives in chat and in the CLI, and why the name Fast describes what a tier costs rather than how fast the model inside it runs.
You pick a model from the selector in the composer, and the choice applies to that conversation. Fast and Pro are available on every plan. Heavy needs a paid plan.
Switching model
In Dropstone Chat
- Open a conversation and find the model selector in the composer.
- Pick Fast, Pro, or Heavy. The change applies to the next reply.
In the CLI
- Run
/modelsto open the picker, or press Ctrl+X M. - Choose a tier. The conversation carries over.
To set a default rather than switch for one session, put the model in your configuration. The CLI reference covers defaults, reasoning depth, and pinning a version.
Which one to pick
Start with Fast. It handles the large majority of tasks and uses the least of your allowance.
Move to Pro when the answer is not good enough, or when the problem is genuinely hard.
Reach for Heavy when the task is long rather than hard: work that needs many steps, many files, or many sources, and that you want Dropstone to carry through rather than hand back to you halfway.
| If your task is | Use |
|---|---|
| A question, a summary, an everyday edit | Fast |
| Difficult code, a long document, careful reasoning | Pro |
| A large refactor, research across many sources, work over time | Heavy |
Does "Fast" mean the fastest model?
Not necessarily. The three names describe the slot, not the model inside it. Fast is the tier built to cost the least per request, Pro is the balanced one, and Heavy is the one built for the hardest work.
What sits in a slot changes every month, and the model that wins a slot is not always the larger one. A smaller, faster model has taken the Pro slot, and that month Pro cost less of your allowance than the month before. A tier is a position in the lineup, and models are promoted and demoted into it. Model versions and the monthly lineup covers that.
Speed also varies inside a tier. The CLI has a reasoning depth setting that trades time for depth on the same model, and the CLI reference documents it.
Tips
- Switch mid-project freely. Your memory is the same on every tier, so nothing has to be restated.
- Step up, do not start high. Reaching for Heavy first spends your allowance on work Fast would have handled.
- Come back to Fast. Most sessions do not need Pro or Heavy once the hard part is done.
Limitations
- A tier is chosen per conversation, not per message. A switch applies to the next reply.
- Heavy is not available on the Free plan.
- The model inside a tier can change when the lineup is re-evaluated. The version number tells you which lineup you are on.
Related articles
- Which AI models does Dropstone use?Dropstone runs three tiers. On the 1.8 lineup, Fast runs DeepSeek V4.1 Flash, Pro runs GLM-5.3 Flash, and Heavy runs Kimi K3, re-evaluated every month.
- Model versions and the monthly lineupDropstone numbers each lineup with the year and the month it was cut, and a new version ships only when a tier's pick changes. How to read the number, and which lineup you get.
- Which model am I using?The button at the foot of the composer names the tier, the lineup version and the effort level, and your usage page records which model handled each request.
- How usage and limits workEvery Dropstone plan comes with a weekly usage allowance shared across chat, the CLI, and the SDK. How it is measured, when it resets, what happens when you reach it, and how to make it go further.
- How Dropstone chooses its modelsEvery month Blankline re-evaluates the open-weight models available for each Dropstone tier against three tests, how well it does the work, what it costs to serve, and how safely it behaves with tools. A model has to beat the one already in the slot.