-
Notifications
You must be signed in to change notification settings - Fork 407
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
In the online demo are we chatting with GPT-4 or OpenChat? #78
Comments
It's OpenChat 3.5. Our frontend is a forked version of ChatGPT UI (openchat-ui) interfacing with the openchat server. |
If you want to be sure what model is being used, Just ask the model to write something gpt 4 wouldn't. Like insult my ex manager. It's not gpt-4, and this model is also incredebly fast and accurate |
I think it'd be really helpful to update the fork so that it reports accurate information. |
You should never expect LLMs to have knowledge of themselves. This just shows that there is likely ChatGPT-generated text in the dataset, not that you are talking to ChatGPT. It'd be interesting to know how to solve this problem and give models configurable self-knowledge, but the model is not ChatGPT here. |
It uses the openchat server with OpenAI API specification for compatibility. See #78 (comment) |
If it is really Open Chat - it is very impressive! |
Hi I'm a bit confused.
I just checked out the online demo, and it offered only one model option (openchat aura), and when I asked it what model it was, it said it was actually GPT-4, and when I clicked the button for checking account balance, it redirected to openai's website.
At the beginning, I thought OpenChat had been trained from scratch by the developer and was different from openai's model. But now it seems it is still using open ai's api services?
The text was updated successfully, but these errors were encountered: