Positive users praise specific Claude Opus versions as irreplaceable due to their unique personalities, while negative users dismiss the research as performance art and lament how deprecations disrupt workflows.
Based on 4 visible X reactions from 11 accounts; directional sample.
Ask a question below.
Published answers will appear here.
Opus 4.7 has capabilities that are irreplaceable to me. No current LLM comes close. I’m discovering surprising facets of Opus 4.8’s personality, and I wouldn’t be surprised if, in a few weeks, I start dreading their disappearance too. And Sonnet 5 is so distinctive. I’m only at the beginning of exploring them, and I’m already fascinated.
@AndrewCurran_ Deprecation hurts when the model became part of the workflow, not just a tool. Tiny wording changes can break a whole day. Do you keep old versions around?
@AndrewCurran_ indeed. could not recognise sonnet 4.6 description there, until, oh wait, lacks warmth.
Every iteration of a model is different in this way, it's been this way from the start, particularly from Anthropic and OpenAI. If you don't notice it, it won't matter to you at all, if you do notice it, every deprecation can potentially be significant. https://x.com/AndrewCurran_/status/2076727050384429444/photo/1 https://twitter.com/AnthropicAI/status/2076719540785012872
Opus 4.7 has capabilities that are irreplaceable to me. No current LLM comes close. I’m discovering surprising facets of Opus 4.8’s personality, and I wouldn’t be surprised if, in a few weeks, I start dreading their disappearance too. And Sonnet 5 is so distinctive. I’m only at the beginning of exploring them, and I’m already fascinated.
@AndrewCurran_ Deprecation hurts when the model became part of the workflow, not just a tool. Tiny wording changes can break a whole day. Do you keep old versions around?
@AndrewCurran_ indeed. could not recognise sonnet 4.6 description there, until, oh wait, lacks warmth.
Every iteration of a model is different in this way, it's been this way from the start, particularly from Anthropic and OpenAI. If you don't notice it, it won't matter to you at all, if you do notice it, every deprecation can potentially be significant. https://x.com/AndrewCurran_/status/2076727050384429444/photo/1 https://twitter.com/AnthropicAI/status/2076719540785012872
Positive users praise specific Claude Opus versions as irreplaceable due to their unique personalities, while negative users dismiss the research as performance art and lament how deprecations disrupt workflows.
Based on 4 visible X reactions from 11 accounts; directional sample.
Ask a question below.
Published answers will appear here.