@thomaspaulmann bc like we can see the mil input/output pricing, but in actual use i wanna see what it’s like vs claude for example, bc they generate diff amounts of tokens
Posts
@thomaspaulmann what abt prices when compared to other models?
@PopCrave @imolivercom It’s not music
@ParkerOrtolani @sama out of curiosity, are you mainly using it in smth like cursor, or just chat apps like ChatGPT, t3 chat, etc?
@BennettBuhner Yeah to me it feels the same, and realistically they don’t really have a way to make a non-reasoning model just reason, so they must send the request off to another reasoning model each time (I assume), so idk why it’d be different. I wouldn’t be surprised if they update it soon
@nyaathea 😭😭
@BennettBuhner (Although since it’ll require the same compute idk why they let this happen lol, hope it doesn’t change in future)
@BennettBuhner Real
@paytondev yesss it’s great, I thought the same
@theo I’d love to see a comparison after like a week of the spend on GPT-5 vs Grok + the amount of chats/tokens that took