Anthropic Passes OpenAI at $30B Revenue as Mythos Preview Rattles the Software Sector

From $9B to $30B in Four Months

Anthropic's annualized revenue tripled from roughly $9 billion at the end of 2025 to over $30 billion by early April 2026, putting it ahead of OpenAI's approximately $25 billion run rate for the first time. The growth is almost entirely enterprise-driven: 80% of Anthropic's revenue comes from business customers, compared to OpenAI's more consumer-heavy mix. The number of clients spending over $1 million annually doubled from 500 to over 1,000 in under two months.

Alongside the revenue disclosure, Anthropic announced an expanded partnership with Google and Broadcom to secure approximately 3.5 gigawatts of TPU compute capacity beginning in 2027. AWS remains the primary cloud provider via Amazon's $8 billion commitment, while Google supplies custom TPUs. Training costs remain a fraction of OpenAI's — roughly 4x less by current estimates — which makes the revenue gap even more striking on a margin basis.

Mythos Shakes Up the Market

On April 7, Anthropic unveiled Claude Mythos Preview, a next-generation model that the company described as too capable to release publicly. During internal testing, Mythos autonomously discovered thousands of previously unknown zero-day vulnerabilities across every major operating system and browser — including a 17-year-old remote code execution flaw in FreeBSD that was triaged as CVE-2026-4747.

Rather than a broad rollout, Anthropic launched Project Glasswing, distributing Mythos Preview to over 50 organizations including Amazon, Apple, Microsoft, CrowdStrike, and the Linux Foundation, backed by $100 million in usage credits and $4 million in direct donations to open-source security groups. The message was clear: this model changes the offensive-defensive balance in cybersecurity, and defenders need a head start.

The market read the subtext. By April 9, US software stocks slumped as investors repriced the disruption risk from frontier AI. The same capabilities that make Mythos a defensive asset for Anthropic's partners make it an existential question for companies whose moats depend on software complexity.

The Tender Offer Signal

On April 8, Anthropic completed an employee tender offer at a $350 billion valuation — the same pre-money figure from its February Series G. Investor demand reportedly reached $6 billion, but employees chose to sell far fewer shares than expected. When insiders refuse liquidity at a $350 billion valuation, the signal is that they expect substantially more — likely from an IPO the company is expected to pursue later in 2026.

Adding fuel, Michael Burry posted that "Anthropic is eating Palantir's lunch", pointing to the revenue trajectory and claiming Anthropic captures 73% of new enterprise AI spending by Ramp data. Palantir dropped roughly 7% on the post. Whether or not Burry's thesis is entirely right, the framing helped crystallize a narrative: Anthropic is not just an AI lab — it is becoming the default enterprise AI vendor.

The ANTHROPIC perpetual now trades near $795, implying a total valuation of roughly $795 billion — more than double the $380 billion post-money from the Series G just two months ago. The market is pricing in both the revenue trajectory and the possibility that a public listing will validate it.

Trading on Hyperliquid

Trade ANTHROPIC on Hyperliquid with up to 3x leverage.