Comment by bcherny
2 days ago
Yeah, adherence is a hard problem. It should be feeling much better in newer models, especially Opus 4.5. I generally find that Opus listens to me the first time.
2 days ago
Yeah, adherence is a hard problem. It should be feeling much better in newer models, especially Opus 4.5. I generally find that Opus listens to me the first time.
Have been using Opus 4.5 and can confirm this is how it feels, it just works.
It also works your wallet
Right now Google Antigravity has free Claude Opus 4.5, with pretty decent allowances.
I also use Github Copilot which is just $10/mo. I have to use the official copilot though, if I try to 'hack it' to work in Claude Code it burns thru all the credits too fast.
I am having a LOT of great luck using Minimax M2 in Claude Code, its very cheap, and it works so good.. its close to Sonnet in Claude Code. I use this tool called cc-switch to swap out different models for Claude Code.
Highly recommend Claude Max, but I also want to point out Opus 4.5 is the cheapest Opus has ever been.
(I just learned ChatGPT 5.2 Pro is $168/1mtok. Insanity.)
If you pay for a Claude Max subscription it is the same price as previous models.
Just wait a few months -- AI has been getting more affordable _very_ quickly
I’ve felt that the LLM forgets CLAUDE.md after 4-5 messages. Then, why not reinject CLAUDE.md into the context at the fifth message?
CLAUDE.md should be picked up and injected into every message you send to the model, regardless if it is 1st or 10th message in the same session.