3 ms·
> dumbing down their models, This should be so easy to prove if it were true. Yet there is none of it, just vibes. Still, your other two points are completely
by deaux 6mo ago
> dumbing down their models,
This should be so easy to prove if it were true. Yet there is none of it, just vibes.
Still, your other two points are completely valid. The opaqueness of usage quotas is a scam, within a single month for a single model it can differ by more than 2x. And this indeed has been proven.
- greenavocado 6mo ago> This should be so easy to prove if it were true. https://github.com/anthropics/claude-code/issues/42796 https://github.com/anthropics/claude-code/issues/42796 https://scortier.substack.com/p/claude-code-drama-6852-sessions-prove https://scortier.substack.com/p/claude-code-drama-6852-sessi...
- deaux 6mo agoFirst link is about the harness, Claude Code, defaulting to less thinking over time. This isn't "the model getting worse". Second link is just a discussion of the first link.