OpenAI is winning major coding customers away from Anthropic with steep discounts, The Information reported on October 24–25, 2026 — the clearest sign yet that the frontier-model price war has moved from list prices to negotiated enterprise deals. The headline case: Replit shifted most of its AI spending to OpenAI after being offered more than 30% off a contract worth over $100 million a year, after Anthropic declined to extend Replit's prior discount without a higher spending commitment.

The numbers at Rippling tell the same story in shares rather than dollars. CEO Parker Conrad said OpenAI's share of the company's coding workload rose from 16% in May to 52%, while Anthropic's fell from 57% to 18% over the same period. DataCamp CEO Jonathan Cornelissen said OpenAI's pricing cut his costs by 80% to 90%. And a US bank with over $7 billion in assets reportedly doubled its ChatGPT seats to 500 employees at no extra charge, paying under $7 per user against a standard $20 or more.

The Replit split is messier than the headline suggests — worth noting honestly. CEO Amjad Masad said at TechCrunch Disrupt that Replit is "no longer working with Anthropic," then walked that back within a day; Replit's own catalog still lists seven Claude models, and OpenAI is now the default provider for Replit Agent while customers who want Claude must pick it. The direction of travel is not in doubt, but "switched entirely" would overstate it.

The timing stings for Anthropic, which is reportedly heading toward a public listing: an investor memo cited in the same reporting says its growth slowed from the second quarter to the third. OpenAI, meanwhile, is partly buying market share at the expense of its own margins — though its cheaper Sol and Luna model lines give it real efficiency gains behind the discounts. Both labs also face the same pressure from below: lower-cost Chinese models from DeepSeek and Kimi. For developers, the lesson is blunt — at this scale, price now decides, and discounts this large are negotiable.

Sources