Censorship (Safety filters) setting on API

[ujayden]

Hi, I noticed that there is a safety filter on the APIs, as mentioned in Google's documentation: Safety Settings.

Could I know if there’s a way to disable the safety settings in the configuration, or I need to modify the code in the proxy to disable the safety filter?

Which part of the code should I edit if I need to do the method 2?

Thanks!

[johnd0e]

Could I know if there’s a way to disable the safety settings

It is already done:

openai-gemini/src/worker.mjs

Lines 188 to 198 in 6559eaf

	const harmCategory = [
	"HARM_CATEGORY_HATE_SPEECH",
	"HARM_CATEGORY_SEXUALLY_EXPLICIT",
	"HARM_CATEGORY_DANGEROUS_CONTENT",
	"HARM_CATEGORY_HARASSMENT",
	"HARM_CATEGORY_CIVIC_INTEGRITY",
	];
	const safetySettings = harmCategory.map(category => ({
	category,
	threshold: "BLOCK_NONE",
	}));

Also AFAIK the new models have them not enabled by default.