Anthropic has achieved a significant milestone in artificial intelligence development with its Opus 4.5 model, which becomes the first system to surpass 80 percent accuracy on SWE-Bench verified coding assessments while launching integrations for Chrome browsers and Excel spreadsheets. The company positions this release as more than incremental progress, targeting enterprise productivity through workplace tools that embed AI assistance directly into existing professional workflows rather than requiring separate applications.
The model demonstrates superior performance across multiple testing frameworks beyond coding benchmarks, handling complex tool usage scenarios, and general problem-solving tasks. Anthropic has addressed longstanding user frustrations by enabling continuous conversations through memory compression technology, eliminating previous context limitations that disrupted extended interactions. The launch completes the 4.5 series following earlier Sonnet and Haiku variants, establishing a tiered system where enterprises can select appropriate models for specific requirements.
Chrome integration reaches all Max subscribers, while Excel functionality targets business customers across Max, Team, and Enterprise access levels, directly challenging recent releases from OpenAI and Google in an intensifying competition for corporate AI adoption.
The model demonstrates superior performance across multiple testing frameworks beyond coding benchmarks, handling complex tool usage scenarios, and general problem-solving tasks. Anthropic has addressed longstanding user frustrations by enabling continuous conversations through memory compression technology, eliminating previous context limitations that disrupted extended interactions. The launch completes the 4.5 series following earlier Sonnet and Haiku variants, establishing a tiered system where enterprises can select appropriate models for specific requirements.
Chrome integration reaches all Max subscribers, while Excel functionality targets business customers across Max, Team, and Enterprise access levels, directly challenging recent releases from OpenAI and Google in an intensifying competition for corporate AI adoption.
