yup. it absolutely does follow the instructions to the letter. still testing its coding abilities. I noticed cursor says, “Model does not support images”. Is this true? I have found it to be one of the best at reading images when I use it on gemini.google.com. Does that mean there is no image support with the API? Gotta to go check the docs.
gemini-2.0-flash-exp, gemini-2.0-flash-thinking-exp, and gemini-exp-1206 are now available for testing in the editor! They’re new and experimental, so excuse some weird behaviour early on, while we work on bringing these models up to the same polish as the Claude and 4o models.
I’ve seen comments on Reddit that gemini-exp-1206 is superior to gemini-2.0-flash-exp and gemini-2.0-flash-thinking-exp in coding tasks, do you guys have the same experience?