Anthropic revenue hits $30B ARR, overtakes OpenAI at $25B; enterprise focus drives efficiency gap

Anthropic's annualized revenue has crossed $30 billion, surpassing OpenAI's $25 billion run-rate and marking a structural shift in AI lab business models. The achievement occurred ahead of analyst projections by several months, with Anthropic hitting the $30B milestone in April 2026—a 30x increase from $1 billion in January 2025 over a 15-month span. OpenAI's revenue grew from $13B in 2025 to $25B by February 2026, a 1.9x increase over the same window, signaling diverging growth trajectories between the two labs.

The revenue composition reveals the strategic divide: Anthropic's revenue is approximately 85% enterprise and developer API usage, with over 1,000 enterprise customers now spending more than $1 million per year on Claude. OpenAI's mix skews heavily consumer, with roughly 85% tied to ChatGPT subscriptions and 95% of free-tier users generating inference costs without offsetting revenue. The gap is amplified by unit economics—Anthropic projects positive free cash flow by 2027 and is on track to reach 77% gross margins by 2028, while OpenAI projects $14 billion in losses for 2026 and doesn't expect profitability until 2029 or 2030.

For AI practitioners and infrastructure planners, the revenue crossover signals that enterprise willingness to pay for Claude API usage—driven by measurable productivity gains in code generation, data analysis, and knowledge work—is outpacing consumer chatbot demand. Anthropic's spending projection of ~$30 billion on training and inference compute through 2030 versus OpenAI's $125 billion for the same period underscores the cost-efficiency advantage of API-first, enterprise-focused development. Both companies are moving toward public markets, but Anthropic's path points toward profitability; OpenAI's to continued heavy infrastructure spending.