Skip to content

Choose providers and models

This guide gets one provider ready and selects the model your agents use. It assumes SASE is installed; if not, start with Install and set up SASE.

Before you start

You need one supported agent CLI installed and authenticated for production work. The bundled fakey provider is for tests, not real runs.

1. Install and authenticate one provider

Each supported provider has an install command, an authentication command, and a link to its vendor docs in Agent providers. SASE can install most provider CLIs for you; the page says which one you install yourself.

2. Check that SASE can use it

Run the authoritative readiness check:

sase doctor -C llm.auth -v

The check prints the selected provider, how it was selected, and the authentication evidence it found. If the page and the doctor ever disagree, trust the doctor output.

A provider reported as disabled is routing policy, not a missing install: a hard disable blocks automatic routing and explicit requests until it expires or is cleared, while a soft disable only steers preference pools away. Interactive provider windows still launch while hard-disabled.

3. Pick the model

Launches without an explicit request use llm_provider.default_model, which is the @large size alias unless changed. The five size aliases — @xsmall, @small, @medium, @large, @xlarge — resolve to an available provider and model, so prompts that use them stay portable across provider choices. One prompt can request something else with a %model directive such as %model:codex/gpt-6-luna.

Set the default in configuration when most of your work wants a different level:

llm_provider:
  default_model: "@medium"

Confirm the new default with sase doctor -C llm.default.

If it goes wrong

Re-run the check from step 2 and follow its next step; provider sections in Agent providers match the doctor hints. For anything else, start with Diagnose problems with sase doctor.

_[project]: A sase project is a named unit of work registered with SASE. _[workspace]: A sase workspace is a numbered clone of a project's primary repo, managed by the workspace store and tracked in that project's registry.json. _[agent]: A sase agent is an agent session or a single agent that does not belong to a session. _[prompt]: A prompt is the text SASE hands to a provider to start an agent turn. _[macro]: A macro is a reusable prompt triggered with #name in agent prompts. _[Patch]: A Patch is SASE's local unit of change. _[bead]: A bead is one tracked work item in SASE's git-native issue tracker. _[plan]: A plan is a persisted, approved artifact recording the intent behind agent work. _[TUI]: The TUI is SASE's terminal interface for running agents and reviewing results. _[epic]: An epic is an executable multi-phase plan run with sase bead work. _[gate]: A sase gate is a durable, command-backed request for one user decision, created when work must pause for the user. _[monitor]: A sase monitor is a monitor turn that runs one long command under a detached supervisor, so the command outlives the agent that started it. *[stitch]: A stitch is the lightweight ordered change record inside a Patch's STITCHES: section.