Is it possible to use Gemini 2 flash, everybody is raving about it and if so does it add more the Cody’s context memory which is limited to 200k, more than other systems yet still too little especially when it keeps dropping the context after a little while and you have to give it some of your files again and again.
Hey @GevPwr
Gemini Flash 2.0 is already built in as an experimental model. If you want higher token limit then you can use your own API key from Google and configure the Cody extension settings as follows:
cody.dev.models: [
{
"provider": "google",
"model": "gemini-2.0-flash-exp",
"inputTokens": 1048576,
"outputTokens": 8192,
"apiKey": "YOUR_OWN_API_KEY",
"options": {
"temperature": 0.0
}
},
]
Happy Christmas
1 Like