Member-only story
A new model, Gemini 3.6 Flash, which cuts paid output pricing from $9.00 to $7.50 per million tokens versus Gemini 3.5 Flash, while its faster as well, Artificial Analysis measures 275.5 output tokens per second instead of 175.7. But is that worth anything?
💥 Master Any Skills in 3 Months📚 Up to50% OFFPremium Courses
⏰ Limited-Time Offer👉Enroll Now & Start Learning
It depends, for high-volume API work, it is a sensible upgrade. However, for us coders that are waiting for a Google model that can stay useful through a messy, long-running coding task, it is not the release we needed. The checked public score is unchanged: Artificial Analysis gives both Gemini 3.5 Flash and Gemini 3.6 Flash an Intelligence Index of 50.
For long-running coding, I think this is a bad release. A very bad release even. Gemini 3.5 Flash already did not earn my trust for serious repo work, and Gemini 3.6 Flash does not provide the result that would make me revisit that judgement. The launch improves the economics of the same tier. It does not demonstrate that Google has solved repo understanding, tool discipline, recovery after a wrong turn, or the…