Anthropic has called off its plans to acquire Decart, an Israeli AI startup, according to a report from calcalistech.com. The deal, which had been under discussion for several weeks, will not move forward, ending what had been one of the more closely watched potential acquisitions in the AI sector this year.
What We Know About the Collapsed Deal
Details on why the acquisition fell apart remain limited. Neither Anthropic nor Decart has issued a public statement explaining the breakdown. When Anthropic's talks to acquire Decart were first reported by Reuters, the deal attracted immediate attention given Decart's reputation for building high-performance AI inference technology. The startup had positioned itself as a specialist in running large language models faster and at lower cost, a capability with obvious appeal to a company like Anthropic.
Key Facts
- Anthropic has ended acquisition discussions with Israeli AI startup Decart
- The deal had been reported at a potential valuation of around $6 billion
- Decart specializes in high-speed AI model inference technology
- Neither company has commented publicly on the breakdown
- The collapse follows weeks of reported negotiations
The valuation figures attached to the deal were striking. Reports had placed Decart's acquisition value at roughly $6 billion, a number that reflected both the premium on inference efficiency expertise and the competitive intensity among frontier AI labs. Whether disagreements over price, structure, or strategic fit contributed to the breakdown is not yet clear.
Anthropic has dropped its plans to acquire Israeli AI startup Decart.calcalistech.com
Context: Anthropic's Broader Expansion Push
The failed deal comes at a busy period for Anthropic. The company has been scaling aggressively on multiple fronts, from compute infrastructure to new product verticals. Earlier this year, Anthropic secured a substantial compute agreement to support its growing model workloads, and it has been building out specialized offerings aimed at sectors including life sciences. Acquisitions would fit naturally into that expansion strategy, making the Decart talks logical even if they ultimately did not close.
Decart is unlikely to remain independent for long given the level of interest it has already attracted. Inference optimization is a crowded and competitive space, with demand coming from every major AI lab and cloud provider. Whether Decart returns to the table with Anthropic or finds a different buyer will be worth watching over the coming months.
For Anthropic, the immediate question is whether the company pursues alternative acquisition targets to fill whatever gap Decart was meant to address. The company has shown a clear appetite for growth beyond its core model development work, and the infrastructure and tooling layer around large language models remains an attractive area for investment. Walking away from one deal does not signal a retreat from that broader strategy.
As the AI industry continues to consolidate, failed deals like this one are becoming more common. Valuations are high, competition is fierce, and the gap between initial interest and a signed agreement can be substantial. This deal joins a growing list of high-profile AI acquisition talks that have not reached a conclusion.
