I think Grok would be great if it had the same quota as Claude. Compared to GPT, it has far fewer unnecessary guardrails and it's fast — it's just that Grok's quota is a bit tight.
@thsottiaux So what if it's $200? I'd gladly pay it if the intelligence never degraded. The real issue is that it keeps getting dumber — worse at getting actual work done, yet still stuck in over-defensive programming.
GPT Luna max thinking mode: runs a hundred defensive tests before doing anything, then the testing itself modifies your stuff mid-run. It defended against the risk — and edited your files.
@thsottiaux After updating the GPT client to version 26.924.22138, the app has crashed many times. This version is very unstable. Please fix it as soon as possible.
@thsottiaux Hey, just wondering — have the conversation compression settings been changed? I’ve noticed it starts compressing after only a few messages.
@thsottiaux I hope the model isn’t only great right after launch, then starts getting dumber and laggy just a few days later. The initial performance is really impressive — it would be even better if that quality could last.
@thsottiaux Selected model is at capacity. Please try a different model.
Paying Pro users being locked out with a “full” excuse is morally indefensible. Taking money then denying service is a breach of fairness.