From the Court Documents Anthropic Admits:
Max 5x email: “Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits.”
Max 20x email: “Most Max 20x users can expect 240-480 hours of Sonnet 4 and 24-40 hours of Opus 4 within their weekly rate limits.”
https://t.co/8g51kegN2Z
From the Court Documents Anthropic Admits:
Max 5x email: “Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits.”
Max 20x email: “Most Max 20x users can expect 240-480 hours of Sonnet 4 and 24-40 hours of Opus 4 within their weekly rate limits.”
https://t.co/8g51kegN2Z
From the Court Documents Anthropic Admits:
Max 5x email: “Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits.”
Max 20x email: “Most Max 20x users can expect 240-480 hours of Sonnet 4 and 24-40 hours of Opus 4 within their weekly rate limits.”
https://t.co/8g51kegN2Z
@TokenGremlin I'll tell you one thing, OpenAI is not the one getting a Class Action Lawsuit for scamming customers out of money. https://t.co/E1ptIc9VPt
From the Court Documents Anthropic Admits:
Max 5x email: “Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits.”
Max 20x email: “Most Max 20x users can expect 240-480 hours of Sonnet 4 and 24-40 hours of Opus 4 within their weekly rate limits.”
https://t.co/8g51kegN2Z
I am pleased to see that Ant’s new Claude Code has a version of background computer use on par with the version from Codex from last May this year. Nice work!
Shipping great features first turns out to be a great way to encourage other labs to ship too. We will continue to do this until other labs pay more attention to shipping. This is good for everyone and there is a lot of room left to go!
We solved computer use in practice for GPT models about four months ago. But computer use is a significantly awesome thing no matter what model you use, and it is important that the industry similarly spends more effort to train their models to be great at computer use, among other elements of model capabilities.
As models become more capable and central to businesses and economies, the value of computer use only increases. We should be taking it seriously, and doing the right thing for our customers and the world.
From the Court Documents Anthropic Admits:
Max 5x email: “Most Max 5x users can expect 140-280 hours of Sonnet 4 and 15-35 hours of Opus 4 within their weekly rate limits.”
Max 20x email: “Most Max 20x users can expect 240-480 hours of Sonnet 4 and 24-40 hours of Opus 4 within their weekly rate limits.”
https://t.co/8g51kegN2Z
Anthropic’s $200 “20x” plan is a scam.
They put 20x on the checkout page so you think you’re getting 4x the $100 plan.
You’re not.
That 20x only applies to a 5 hour window. One sitting. Then it resets and they hope you don’t notice the real cap.
The weekly limit is what actually kills you.
$100 plan: 15–35 hours of Opus a week
$200 plan: 24–40 hours of Opus a week
You pay DOUBLE and get maybe 10 extra hours of the model you upgraded for.
Sonnet weekly is 140–280 vs 240–480. That’s 1.7x. Not 4x. Not 20x.
Once the weekly cap is gone the 5 hour reset does nothing. You’re locked till next week. They never published real numbers. People had to go dig through Reddit AFTER they already paid $200. Now they’re getting sued because the $200 plan is more like 6–8x Pro, not 20x.
This is the only thing Scamthropic is good at.
Codex 20x is actually 20x of Plus on the weekly limit. No 5 hour cap on the Pro plans. The numbers are on the page and they multiply like the label says. Tibo just said it out loud.
That’s the difference.
https://t.co/Z4dTv7kFat
I resigned from Anthropic today. I spent the last three years doing pretraining research at both OpenAI and Anthropic. Neither company is acting responsibly. They are racing straight to self-improving superintelligence and gambling with our lives. More thoughts below.