Grok CLI guide: models and effort

Grok CLI models and the context window

Grok is the straightforward one on context. It reports the real window rather than an advertised figure you have to correct, and it compacts at a fixed proportion of it. The model list arrives in the opening handshake rather than from a separate call.

xAI

On Grok CLI

The window is the real one

Grok 4.5 reports a 500,000 token context window and that is the number you get. Unlike Codex there is no harness reservation to subtract, so a usage bar can use the reported figure directly.

Compaction is a threshold, not a smaller window

Auto compaction fires at 85 percent of the window. That is a point at which something happens, not a reduction in how much context you have, and treating it as a smaller denominator makes every usage readout wrong.

The model list arrives at handshake

Grok reports its available models as part of the initial handshake rather than through a separate catalog call. A client already has the list before the first turn.

Effort levels are per model

As with Codex, the levels belong to the model rather than to a fixed list, and they come from the same handshake.

Variants inherit their base

An agent variant of a model, something like a build-specific name, generally shares the context window of the model it is based on. Its thinking support is a different matter and should not be assumed from the name.

Compare

Model and reasoning controls by engine

 Claude CodeCodex CLIGrok CLI
Pick a modelYesYesYes
Reasoning effortYesYes, per modelYes, per model
Separate thinking budgetYesNoNo
Where the level list comes fromThe engineThe model catalogThe initialize handshake
Reports its context windowYesYes, larger than usableYes, and it is accurate
Auto compactionYesYesAt 85 percent of the window

Termdeck drives all three from one console, so the differences above stop being three separate habits to remember.

FAQ

Grok CLI models and effort, answered

How large is the Grok CLI context window?

Grok 4.5 reports 500,000 tokens, and that is the usable figure rather than an advertised one. There is no harness reservation to subtract, so the reported number can be used directly.

What happens when a Grok session fills its context?

Auto compaction fires at 85 percent of the window. That is a threshold at which the conversation is compacted, not a smaller window, so the full 500,000 tokens remain the denominator for any usage readout.

Where does Grok get its model list from?

The opening handshake. Grok reports its available models when a client connects rather than through a separate catalog request, so the list is in hand before the first turn runs.

Do Grok agent variants have the same context window?

Usually yes, because a variant generally inherits the window of the model it is based on. Whether it supports the same thinking controls is a separate question and should not be inferred from the name.