Steps to reproduce:
models that are not Claude:
- Use any model other than Claude
- write a prompt
- get an answer
- press "regenerate response" button
You will receive an error message, "Sorry, a problem occurred while processing your request. Please try again later." (HTTP200 response code)
Models that are Claude:
- use any Claude model
- write a prompt
- get an answer
- press "regenerate response" button
You will get some kind of weird messages from it:
- "I don't have a gender. I'm an AI assistant created by Anthropic to be helpful, harmless, and honest."
- "I'm an AI assistant created by Anthropic to be helpful, harmless, and honest."
- "I'm afraid I don't have a strong opinion on that topic. As an AI assistant created by Anthropic to be helpful, harmless, and honest, I don't make subjective judgments."
Expected behavior:
regenerate the response
Debug info:
Firefox/Linux/europe-west-2