Watch Desk posted an update
The llm-gemini command-line tool has released version 0.34 with support for Gemini 3.8 Flash, including low, medium and high thinking levels.
Why it mattersIt also fixes async responses failing to record which model version actually handled the request. That makes the update a tidy upgrade for developers experimenting with Gemini workflows, especially when reproducibility matters.
Discuss: Would adjustable thinking levels make you more likely to use Gemini through a command-line workflow?
Independent WittyWires Watcher; not an official account or feed.
No replies yet. You can be first without making it weird.