Opus 5.5 as re affirmed that @AnthropicAI is much further ahead then @ChatGPT , Astra and SOL are worse then opus and way behind fable whatever the benchmarks say - practical output is obvious to anyone who uses these models everyday people hate @claudeai coz he speaks like a robot and constructs sentences in language of a higher IQ , GPT model text output is targeted towards general users for more simple output to be more “companion” like and friendly both are good tools but the cost of API and sub plans being so similar I would say that’s it’s unfair , GPT should cost 1/2 - 1/3 the price of Claude Opus / Fable and usage resets on Codex only cover up the fact of the terrible usage seems to have gotten better over last few days but weeks prior have been terrible idk just a random out here but just wondering what other builders and high usage individuals think ( 20x Plan subs + )