This is likely to be inspect element text editing. But if this is real, then this is a total narrative violation of Chinese LLMs not being "good enough" and will completely explain why Anthropic is trying to scare everyone to use their services.
This is a damning leak if true and absolutely no-one should believe what Dario says at this point.
HN seems to redirect to old.reddit, which doesn't seem to show images on my phone at least, and the author included an image [1] of the model selection in one of the subsequent comments. Anybody able to verify the claim?
3 comments
[ 4.1 ms ] story [ 17.5 ms ] threadThis is a damning leak if true and absolutely no-one should believe what Dario says at this point.
[1] https://preview.redd.it/anthropics-claude-remote-uses-glm-4-...