Comment by gosukiwi
4 days ago
Claude and Codex pricing will eventually have to come down, for most common coding tasks you don't need a super smart slow model but a smart-enough and very fast one.
4 days ago
Claude and Codex pricing will eventually have to come down, for most common coding tasks you don't need a super smart slow model but a smart-enough and very fast one.
I’m not convinced they can come down, especially as they both are opening their books for S1 IPO filing.
cheap token for the win.
Microsoft github copilot recently changed their billing. i'm on the yearly subscription. GPT-5.4 is now 6x and even previously free model like GPT-5 mini now cost .33x. its only June 11 and my usage is now at 50%.
I don't think many understand that Sonnet and even Haiku can probably accomplish their task, instead of them invoking a beast like Opus to tell them about todays weather.
And yet, MiMo and DeepSeek, even MiniMax, are way cheaper and arguably better, or way better than both Sonnet and especially Haiku.
While you can argue you are ready to pay 100-1000 times the price for Fable or Opus because you need those last 1-2% of edge, there's no valid reason to keep paying the obscene amounts of money for Sonnet and Haiku when alternatives exist.
I think that the reason is that American labs only focus on their biggest models, and make the smaller ones afterthoughts distilled from the main models. I wonder what would happen if they decided to optimize and train better smaller models.
2 replies →
I don't known how Codex works, but we can set environment variables and point Claude CLI to deepseek. I think that before slashing prices they will slash those environment variables. After all they are not working to give a free TUI to deepseek and possibly to other competitors. But eventually yes, prices will go down or there will be an attempt at a regulatory capture.
Claude Code TUI is garbage. There's nothing worth protecting in there.
Honest question: what's garbage in that tool? I like that I can pair it with whatever editor I am using even if for 99% of the time I run it from inside emacs. I usually chat in the Claude window and it edits my files. I can also write a comment in the code, press a key and emacs will tell Claude to implement that comment. Simple needs, simple tool.
Most importantly, we need a model that doesn't randomly refuse us when we ask it to do something, or worse, deliberately sabotages us when it thinks we're building competing products. Like Anthropic's Fable.
[flagged]