Hello, I’m currently trying to understand the quota / RPM limits for gemini-2.5-flash-image models. When I used gemini-2.5-flash-image in Filter inside Vertax AI API under API/Service Details view, I can see a variation of the model. There are gemini-2.5-flash-image-ga, gemini-2.5-flash-image-ga_default_res, gemini-2.5-flash-image-ga_high_res, gemini-2.5-flash-image-ga_low_res.
I couldn’t find any info regarding the differences between those models and which one gets chosen when I send a request like below (based on google-genai package).
resp = await client.aio.models.generate_content(
model='gemini-2.5-flash-image-001',
contents=contents,
config=generate_content_config,
)
Since each variation has different RPM value, it’s difficult to understand which one should be used. Could you explain the differences and exactly which RPM limit value should be used when I am using gemini-2.5-flash-image-001 for the model name?