Privacy?

[vic-t]

I kept running into issues with continue.dev, so I decided to give kilocode a go. A first quick test looked quite promising.

It could be that I just missed them but I couldn't find any settings with regards to privacy. Specifically, I usually don't agree to model training based on my data.

An official statement on this would be appreciated.

[Juice10]

Hey Vic-t, DevRel Lead at Kilo Code here. We don't ever train on your data.

Since we do offer support for as many models as possible, depending on the model you choose, they might train on your data. Currently I don't have a nice overview for you to see which ones do and don't on our website, but the easiest way to check this for yourself is to go to the OpenRouter.ai website select the model you'd like to use and hover over the document icon.

Here is an example of one that does:

and doesn't train on your data:

Hope this answers your question. I do think it worth it for us to make this more visible somehow if we can.

[vic-t]

Hi @Juice10, thanks for replying so quickly.

OpenRouter seems to have a setting that mitigates the issue. From the documentation:

On your account settings page, you can set whether you would like to allow routing to providers that may train on your data (according to their own policies). There are separate settings for paid and free models.

It would be great if you could take this into account at Kilo Code and make this setting available to us.