[KongchangAI]
Model

gemini-3.5-flash

A lightweight large language model released by Google and part of the Gemini 3.5 series, designed for fast, low-cost inference tasks and suited to scenarios such as simple classification and formatting.

Timeline (last 90 days)

Oct 16

In the author's test, calling gemini-3.5-flash consecutively succeeded on the 1st through 7th attempts and failed on the 8th at 8.4 seconds, with the error giving a retry delay of 46 seconds, and the actual number of allowed requests exceeded the limit value of 5.

Expired50%

Source Articles