Howdy all,
New to these forums. Apologies if this isn’t the right place to ask.
Are there any plans/roadmaps for future Computer Use models? The gemini-2.5-computer-use-preview-10-2025 is pretty outdated (by AI standards) and gemini-3-flash-preview is constantly overwhelmed with 503 errors. On gemini-3-flash-preview specifically I have been experiencing <50% success rates.
This all aside, I love the Gemini API as a tool, and how it extends the potential of my projects. I’m just an amateur, but using the Gemini API makes me feel much more empowered to achieve my goals. The new gemini-3.5-flash model specifically has been excellent at automated reconciliation of my database with information from external API’s.