Published

Technology and Science17 October, 20262 min read

Replit CEO Amjad Masad Says Company Stops Working With Anthropic

This article is an automated aggregation of coverage from news outlets. Our editorials offer in-depth analysis curated by the NewsCord team.

Replit stopped working with Anthropic for months, Masad says at TechCrunch Disrupt. Cost pressures drove the decision, despite an August partnership with OpenAI.

Replit CEO Amjad Masad Says Company Stops Working With Anthropic
Image: TechCrunch

Replit ends Anthropic

Replit CEO and co-founder Amjad Masad said onstage at TechCrunch Disrupt that Replit is no longer working with Anthropic. Masad framed the decision around cost, partnership, and performance, and he said, "I think ChatGPT is better now." Masad said Replit uses an internal benchmark to assess metrics like performance and also publishes a public benchmark.

Masad said Replit consistently sees "Sol and Astra" perform better on building vibe-coded applications from scratch. Masad also said Replit sells a cloud-based software development platform where users can develop and deploy apps directly from inside their browser.

Image from TechCrunch
Image: TechCrunch
Reporting for this sectionTechCrunch

Partnership dispute

A TechCrunch source familiar with the Replit-Anthropic partnership said the cooperation is still ongoing, despite Masad's onstage claim. TechCrunch reported that Replit announced a partnership with OpenAI in August, but there was never any indication that it was no longer with Anthropic before Masad spoke. TechCrunch pointed to evidence on Replit's website, blog posts, and social media posts that Replit was using Anthropic products as recently as a few weeks ago.

TechCrunch cited a blog post in late September that cited Anthropic’s Claude Fable 5 and Claude Fable 5.1 as models running in Replit Agent’s production environment. TechCrunch also noted that one of Replit’s product employees posted on X about using Anthropic’s flagship model Opus 5.5 and promoting its capabilities.

Reporting for this sectionTechCrunch

Benchmarks and safety

Masad said Replit evaluates models using an internal performance benchmark and also publishes a public benchmark. Masad described an "agent habitat" for safe deployment that includes orchestration, error recovery, good safety and security and sandboxing, good networking, and identity and authorization. Masad said once that habitat is in place, Replit can deploy AI safely, and he said, "then you can deploy AI safely."

In the same session, Masad said enterprise was like 2% of revenue last year and that this year enterprise is like 15% of revenue, while more than 50% of revenue is B2B. Masad also said Replit agents can spin up databases and key-value stores inside a production environment and that Replit can deploy its entire stack into government and enterprise environments as an on-premises solution.

Reporting for this sectionTechCrunchUnite.AI