Anthropic launched Claude Opus 5.5 with 40% lower typical running costs and faster output while reports tie the company to a November IPO near $2 trillion.
Key Points:
- Anthropic says Opus 5.5 costs about 40% less than Opus 5 on typical token-billed workloads.
- The model scored 66.4% on Terminal-Bench 4.0 and generates output more than 30% faster.
- A reported November IPO could value Anthropic near $2 trillion, but the company has not confirmed final terms.
Claude Opus Pricing
Anthropic released Opus 5.5 on Sep. 22 as the first model in its Claude 5.5 family, targeting coding, research and professional knowledge work. The company says the model performs near Claude Fable 5.1 on most tasks while using fewer tokens than Opus 5.
API pricing is $4 per million input tokens and $20 per million output tokens, both 20% below Opus 5, while cache reads fell 60% to $0.20 per million tokens.
Anthropic estimates those changes, combined with lower token use, reduce typical running costs by about 40%. Output generation is also more than 30% faster.
Anthropic reported a 66.4% score on Terminal-Bench 4.0, up from 52.3% for Opus 5, along with 57.8% on CursorBench 4.0 and 1,846 on GDPval-AA v2.1. The company also cautioned that small benchmark gaps have become less reliable indicators of real-world differences at current capability levels.
Early testers used Opus 5.5 for a 680,000-line code migration completed in less than a day and a 200,000-line code audit finished in under three hours. Anthropic said Opus 5 needed more than 20 hours for the latter task. Opus 5.5 is available through Claude, Amazon Web Services, Google Cloud and Microsoft Azure, with Sonnet 5.5 and Haiku 5.5 expected in the coming weeks.
Also Read: The US Economy Makes Billions From You Not Bothering, Says Citrini Research
Anthropic IPO Context
The release arrives during a broader dispute over how quickly frontier AI should advance. Anthropic CEO Dario Amodei has urged companies to slow capability gains, and OpenAI CEO Sam Altman and Elon Musk publicly backed parts of his proposal. Donald Trump took a different position on Sep. 22, saying the U.S. would not “stifle growth” in AI and that law enforcement could intervene if needed.
The Wall Street Journal reported that Anthropic is preparing for a November IPO that could raise as much as $100 billion and value the company at about $2 trillion.
Those figures remain provisional, and Anthropic has not announced final offering terms. Lower inference costs could become more important if the company seeks public investors while expanding enterprise use.
The IPO timetable has already shifted. Earlier reporting placed a possible listing in October, while the latest report moves the target to November, allowing Anthropic to include stronger third-quarter financial results. That change underscores that the timing, fundraising target and valuation remain subject to further revision.
Read Next: Trump Renames AI “Super Intelligence” For U.S. Government Documents
