Anthropic Adjusts Subscription Plans: Max Renamed, Ultra's 5-Hour Limit Removed

Smart Things reported on October 31 that today, Anthropic announced adjustments to its subscription plan policy: it will rename the Max 5x plan to Max, and rename the Max 20x plan to Ultra. The Max plan remains priced at $100 per month, and the Ultra plan remains priced at $200 per month.

At the same time, Anthropic will remove the 5-hour usage duration limit for verified Ultra accounts, and this benefit applies to most existing subscribers starting today. New accounts will still retain a temporary usage duration limit until verification is completed, and will receive an email notification once the limit is lifted.

▲Anthropic posted an announcement on social platform X

For many developers, this is equivalent to removing a "headache-inducing" restriction. Previously, even if they purchased Anthropic's highest-tier plan, there was still a usage quota that reset every 5 hours. Many developers, while writing code and debugging solutions, frequently hit the limit midway, forcing their work to be interrupted.

However, the Pro plan at $17 per month and the Max plan at $100 per month will still retain the 5-hour usage limit.

An overseas netizen joked on X: "Fable 5.5 will quickly consume your 5-hour usage limit, and the only solution is to buy the $200 plan."

▲Comments from overseas netizens on X

Tibo, the head of OpenAI Codex and ChatGPT, also appeared in the comments section of Anthropic's announcement, commenting, "Best to announce a 28-day reset period," with an obvious competitive tone.

▲Comment from Tibo, head of OpenAI Codex and ChatGPT

Anthropic's plan adjustment this time can be seen as its latest move to compete for and retain users amid intensifying competition. Just yesterday, OpenAI announced that the subscription usage quotas for Codex and ChatGPT Work's Pro100, 200, 500, and Business Premium would be doubled, releasing more membership benefits to users. The white-hot competition among AI companies has spread from model capability contests to plan benefits and user acquisition.

This also comes on the eve of Anthropic's sprint toward an IPO, but unfavorable signals have emerged regarding credit ratings and financing costs. Previously, on October 26, Anthropic submitted an S-1 registration statement to the U.S. Securities and Exchange Commission (SEC), and foreign media reported that its goal is to go public in the week of November 16. According to the disclosure standards in the S-1, Anthropic's revenue in the first half of 2026 reached $16.286 billion (about RMB 108.598 billion), yet its net loss was as high as $80.748 billion (about RMB 538.444 billion).

According to a report today by the UK's Financial Times, Anthropic is expected to receive a double-B rating from credit agencies after its IPO, the highest level for speculative "junk-grade" bonds.

The report said that investment giant Blackstone Group is an important investor in a recent $63 billion financing deal for Anthropic, aimed at funding the company's chip research and development. Blackstone had previously said that Anthropic would obtain an investment-grade rating before 2027, indicating lower risk for lenders.

However, Blackstone's judgment is only a prediction. If the final rating remains at the high end of speculative grade, it will become more difficult and costly for Anthropic to raise funds for the large-scale spending needed to maintain its business. Over the past 12 months, the company has continued to invest heavily and signed a large number of major computing power orders, which investors are particularly worried about.