search

enricoros / big-AGI

Generative AI suite powered by state-of-the-art models and providing advanced AI/AGI functions. It features AI personas, AGI functions, multi-model chats, text-to-image, voice, response streaming, code highlighting and execution, PDF import, presets for developers, much more. Deploy on-prem or in the cloud.
https://big-agi.com
MIT License
4.89k stars 1.1k forks source link

[BUG]Error updating Gemini model: invalid_enum_value #565

Closed dovegq closed 1 month ago

dovegq commented 1 month ago

Description

I'm encountering an error when attempting to update the Gemini model. I'm receiving the following error message:

[ { "received": "createCachedContent", "code": "invalid_enum_value", "options": [ "countMessageTokens", "countTextTokens", "countTokens", "createTunedModel", "createTunedTextModel", "embedContent", "embedText", "generateAnswer", "generateContent", "generateMessage", "generateText" ], "path": [ "models", 11, "supportedGenerationMethods", 2 ], "message": "Invalid enum value. Expected 'countMessageTokens' | 'countTextTokens' | 'countTokens' | 'createTunedModel' | 'createTunedTextModel' | 'embedContent' | 'embedText' | 'generateAnswer' | 'generateContent' | 'generateMessage' | 'generateText', received 'createCachedContent'" } ]

Device and browser

windows11 + Chrome 125.0.6422.142 + big-AGI 1.16.0

Screenshots and more

image

Willingness to Contribute

enricoros commented 1 month ago

Thanks for reporting @dovegq, you were very quick to find out. Google added a new capability to the Gemini model family and we have a strict protocol checker. Sorry for any inconvenience.

dhcgn commented 1 month ago

Is there a workaround for this issue? I just added the GEMINI_API_KEY and didn't load the models in the past.