BTW, I've basically stopped using Opus entirely and I now have several Codex tabs with GPT-5-high working on different tasks across the 3 codebases (HVM, Bend, Kolmo). Progress has never been so intense. My job now is basically passing well-specified tasks to Codex, and reviewing its outputs. OpenAI isn't paying me and couldn't care less about me. This model is just very good and the fact people can't see it made me realize most of you are probably using chatbots as girlfriends or something other than assisting with complex coding tasks
(sorry Anthropic still love you guys 😢)
PS: I still use Opus for hole-filling in VIM because it is much faster than gpt-5-high there
> be a big AI lab > identify top AI "influencers" > pretrain your next model on their codebases > ? > free marketing!
@sama Since you insist, just 2 minor features would do it for me: - The IMO gold medal reasoning model (tomorrow if possible) - Train it on the λ-calculus, linear types, HVM, Bend, proofs, and how to write code in this format: https://github.com/HigherOrder... Will gladly help with that, ty
@DavidOndrej1 yes
@canalCCore2 nem conheço, pq é bom?
@ddunderfelt that doesn't explain it. sometimes I use very simple prompts and it still delivers. I guess people are starting from poorly written codebases. no AI does well in an insalubrious environment
@AIMachineDream no, o3 and o1 were embarrassingly bad for coding
@erdaltoprak @teknium I don't understand why he doesn't like it. Sometimes I wonder if GPT-5 has been explicitly trained to do very well on Haskell / functional programming stuff (or even my own codebases...)
@anujsyadav01 OpenAI should definitely pay me 😠 Where do I send the invoice
@sxohom not really. are there limits? I think I'm using with the API key