Request to use Gemini-1.5-flash

• Is this a new model to our tech stack: Yes
• Will it be used in production: Potentially - for an internal only feature.
• What is the intended use-case? The current feature is using a model that has high latency. We want to see whether Gemini-1.5-flash will improve latency and performance for this feature as an experiment. (We cannot run evaluations on this feature because it is not mature enough to have evaluations)