IBM shares climbed 4% in premarket trading on Tuesday after the company said it would start adding Anthropic’s Claude AI models to its software tools in a press release.
This will immediately bring Claude into IBM’s new AI-focused coding tool and eventually into other products the company sells to enterprise clients.
The first rollout involves an integrated development environment designed to help engineers write and fix code faster using artificial intelligence. Over 6,000 developers are already testing this setup.
IBM is offering them direct access to Claude, the same AI family that helped drive Anthropic to a $183 billion valuation just last month. The Claude models are being built to handle long-form instructions, repetitive coding tasks, and even operate a user’s computer to take actions on their behalf.
IBM expands AI stack using Claude model from Anthropic
The decision to use Claude came as Anthropic launched an upgraded version of its mid-sized model, called Sonnet 4.5. On Monday, the company said Sonnet 4.5 could run independently for up to 30 hours on coding tasks without human input.
That’s a huge jump from the previous Claude Opus 4 model, which topped out at about seven hours. The new version can also follow detailed instructions better than before and interact with computers to carry out actions directly, a feature that first appeared last year.
Jared Kaplan, who co-founded Anthropic and serves as its Chief Science Officer, said Sonnet 4.5 is “stronger in almost every way” than Opus. He also confirmed that a new version of Opus is already being built and should be ready before the year ends.
“We get benefits from having usage at both model sizes,” Jared said in Monday’s announcement. That mix of small and large models is part of the strategy that’s made Anthropic a top player in AI tools built specifically for coding.
The Claude lineup will now sit inside IBM’s toolset, just as developers across industries try to keep up with automation. One area that’s getting attention is vibe-coding, a term for AI tools that allow anyone, even people with no technical skills, to build simple programs.
This kind of use case is why big tech firms like IBM and others are aggressively locking down AI partnerships. Last week, SAP SE’s stock rose after it revealed a deal with OpenAI to bring its tools to Germany’s public sector in 2026.
AI tools target financial and security firms in next phase
Mike Pizzi, Morgan Stanley’s Global Head of Technology and Operations, said AI has already changed how the bank’s engineers write code. “It’s having a pretty profound impact,” Mike said last week during an internal update.
The financial sector is expected to be a major testing ground for tools like Claude Sonnet 4.5, especially for companies looking to automate internal tech systems without risking downtime.
Anthropic also said its new model is performing better in sectors like cybersecurity and finance, areas where accuracy and task management are critical.
The company hit $5 billion in run-rate revenue back in August, with a chunk of that tied to its software tools aimed at making programming easier for big businesses.
Google and OpenAI are also developing models to help with similar tasks, and competition is heating up as more firms bet their future on AI productivity.
Mike Krieger, Chief Product Officer at Anthropic, said companies need more than just models. “There are a few things that need to happen,” Mike said. “The models need to keep improving, people need to adjust how they work, and there needs to be a deeper level of partnership between some of the frontier labs and these enterprises.”
The IBM and Anthropic partnership comes just one week before OpenAI is expected to host its developer conference, where it’s likely to show off updates to its own coding models.
With pressure building across the industry, IBM is moving early, embedding third-party tools that it believes will keep it in the game while rivals chase similar deals. The company did not say when the Claude models would be added to other product lines beyond the developer IDE.
IBM used the word “accessible” in its statement to describe how it’s giving customers access to Claude, but didn’t mention any licensing structure or price details.
Want your project in front of crypto’s top minds? Feature it in our next industry report, where data meets impact.
Source: https://www.cryptopolitan.com/ibm-stock-surges-4-on-anthropic-ai/
