Google launches Gemini 3.5 Flash low to cut token usage
Google just dropped Gemini 3.5 Flash (Low) on its Antigravity platform, aiming to make AI tasks way more efficient for everyone.
This new model is all about using fewer tokens for simple jobs, something users have been asking for since Google switched up how it counts usage.
After a couple of rate limit increases, Google decided to build this lighter, smarter version.
Google resets Gemini quotas across plans
Varun Mohan from Google DeepMind shared that the Low variant uses about 45% fewer tokens than the Medium one, but still keeps performance strong.
The original Flash is now called Medium, and High remains reserved for tougher tasks.
Plus, Google has reset Gemini quotas across all plans (including free tiers) to help users get more out of their projects and tackle those rate-limit headaches.