Sudo · local chat

Thinking Enter send · Shift+Enter newline · Esc stop

Settings

Server endpoint

Where the model server is listening. Default http://127.0.0.1:8080/v1.

Max response tokens

Qwen3 spends part of this budget on reasoning when Thinking is on. 1024 is a good default.

Thinking mode

Shows the model's reasoning in a collapsible block before the answer. Slower, often better on hard questions.

Skill

—

System prompt

Applied to conversations started after you change it.

Appearance

This session

—