OpenAI Considers Token-Price Cuts as Anthropic Surges

OpenAI Considers Token-Price Cuts as Anthropic Surges

OpenAI is weighing steep token-price cuts to counter Anthropic’s rapid revenue growth; the change could affect OpenAI’s planned IPO.

OpenAI is considering significant reductions to the prices it charges for tokens as it responds to rapid revenue gains at competitor Anthropic. Company discussions on potential price changes were reported on June 10 and are described as still in flux.

Anthropic’s annualized revenue run rate rose from about $1 billion at the start of 2025 to roughly $30 billion by April 2026. Anthropic CEO Dario Amodei said the pace exceeded the company’s forecasts by a factor of eight. The firm’s coding product, Claude Code, reached $1 billion in annualized revenue within six months of its public launch in May 2025 and exceeded $2.5 billion by February 2026. Some analysts estimate Anthropic’s annualized revenue may have reached as much as $47 billion by May 2026.

OpenAI’s revenue was smaller in comparison. The company reported a revenue run rate of about $13 billion in 2025 and does not expect to generate positive free cash flow before 2030. Anthropic projects it will surpass OpenAI’s revenue by 2029 and aims to be free-cash-flow positive by 2028.

Both companies are preparing for public listings. Anthropic completed a $30 billion Series G financing in February 2026 at a $380 billion post-money valuation. OpenAI has discussed a potential funding round that could value the company at $750 billion.

Company executives have discussed using lower prices to increase usage and market share. At the same time, running frontier models requires large infrastructure and compute spending. OpenAI leadership has said training a single competitive model can approach or exceed $1 billion. Lower token prices could raise usage but would not necessarily raise profits if higher volume does not offset slimmer margins.

No final decision on token pricing has been announced. Any change would affect enterprise and developer customers who pay for usage and could prompt pricing or product adjustments across the AI market.

Articles by this author