Skip to content
Dash
Get started for free Get started for free

Choose your AI model

Dash supports multiple AI models. You can switch between them and adjust how deeply Dash thinks before responding.

Available models

Section titled “Available models”

The model picker shows every model your workspace can use. Each one has a label (like “Opus 5.5” or “Sonnet 4.6”) and an availability status:

  • Available — ready to use.
  • Preview — available for testing, may change.
  • Deprecated — still works for existing selections but not recommended for new use.

Opus 5.5 is the default model. New workspaces start on the Free plan with Opus 4.6 at medium thinking effort. After you upgrade, the workspace stays on Opus 4.6 until an admin picks a different model.

The model you pick applies to every conversation in the workspace. Individual members can’t override it.

Switch the model

Section titled “Switch the model”

From Slack or Teams:

@Dash switch to Sonnet

@Dash use Opus

Dash will confirm the switch. It takes effect on the next conversation.

From the web dashboard:

Go to app.dashpup.ai, open Settings, and pick a model from the selector. Only admins can change the model.

Workspaces on the Free plan use a fixed model with medium thinking effort (Opus 4.6 for new workspaces) and cannot change the model or effort. Upgrade to a paid plan to unlock model selection.

Thinking effort and speed

Section titled “Thinking effort and speed”

Each model has two tuning settings:

  • Thinking effort — how deeply Dash reasons before answering. Higher effort means more deliberate, careful responses. Lower effort means faster, more direct answers. Good for simple lookups vs. complex analysis.
  • Speed — every model currently runs at standard speed. The speed setting shows on the Settings page, but there’s no faster option to pick right now.

Thinking effort resets to the model’s default when you switch models. Adjust it from the Settings page on the web dashboard, or ask Dash:

@Dash set thinking effort to high

When to switch models

Section titled “When to switch models”
  • Complex analysis, multi-step reasoning, code review — use a stronger model with higher thinking effort.
  • Quick lookups, simple drafts, routine reports — a faster model with lower thinking effort saves time and cost.
  • Cost-conscious teams — lighter models cost less per message. Check your usage on the dashboard.

What’s next

Section titled “What’s next”