Artificial Intelligence

Replit Makes OpenAI Default Over Anthropic, CEO Says

AI coding platform Replit has shifted its default model provider from Anthropic to OpenAI, according to CEO Amjad Masad.

Radwan AbdulhadiEditor and technology writer

Published: October 18, 2026 · Updated: October 18, 2026 · 3 min read

Replit Makes OpenAI Default Over Anthropic, CEO Says — Image: TechCrunch

The Shift to OpenAI

Replit, the popular AI coding platform utilized by startup founders and developers, has shifted its default model provider from Anthropic to OpenAI. Speaking onstage at TechCrunch Disrupt, CEO and co-founder Amjad Masad clarified that the decision was driven by multiple factors, including cost and overall performance.

Although Replit announced a partnership with OpenAI in August, there was no prior public indication that its relationship with Anthropic had changed. Masad explained that ChatGPT's current capabilities made it a more attractive option for the platform moving forward.

“I think ChatGPT is better now,” Masad said during the event, highlighting that internal metrics and public benchmarks indicated stronger performance in building vibe-coded applications from scratch. He added that the choice came down to cost, partnership dynamics, and performance.

Platform Accessibility and Claude Availability

Despite the shift in default settings, Anthropic has not been completely removed from the ecosystem. Following the event, Masad clarified that users can still access Anthropic models on the platform if they choose to select them manually.

Because most users tend to stick with the default settings rather than manually switching providers, the transition could significantly reduce Replit's spending with Anthropic. Masad later emphasized to TechCrunch that the cloud-based development platform remains model agnostic.

“Our users can choose Anthropic’s Claude or other supported models…OpenAI is now our preferred partner for powering Replit Agent, based on performance, economics, and the strength of our partnership,” Masad stated over the weekend.

Past Collaboration and Industry Reception

Before striking the agreement with OpenAI, Replit depended entirely on Anthropic’s Claude models to power its infrastructure. That previous collaboration was widely credited with helping the company scale its revenue significantly, with Masad noting that Replit had recently ranked among Anthropic's top five customers.

However, Masad's remarks on stage were met with some skepticism from industry insiders familiar with the collaboration. An anonymous source asserted that the partnership with Anthropic remained in place.

Public records across social media posts and company documentation also showed active promotion of Anthropic products as recently as a few weeks prior to the announcement, indicating a rapid evolution in vendor relationships.

Enterprise Growth and AI Safety

Alongside the announcements regarding its model providers, Masad shared key insights into the company's financial growth, highlighting a notable shift toward enterprise adoption.

“We’re not sharing ARR numbers just yet, but in terms of growth, like, I would say, enterprise was like 2% of revenue last year,” Masad stated. He noted that enterprise contracts now account for roughly 15% of total revenue, with B2B sales making up more than half.

Looking ahead, Masad projected that enterprise and government customers will drive over half of Replit's revenue by 2027. He also underscored the critical importance of proper infrastructure and security for deploying artificial intelligence safely.