Anthropic Unveils Claude Opus 4.8 With Major Upgrades for Developers

Anthropic has announced Claude Opus 4.8, the latest version of its flagship AI model that brings improvements across coding, reasoning, knowledge work, and agentic tasks while keeping the same pricing as the previous Opus 4.7 release. The company says the new model acts as a more effective collaborator and delivers stronger performance across a wide range of real-world tasks.

One of the biggest upgrades in Opus 4.8 is improved reliability and judgment when handling agentic workflows. Early testers reported that the model is more likely to acknowledge uncertainty and less likely to make unsupported claims, which helps reduce errors during complex projects. Anthropic’s internal evaluations show that Opus 4.8 is around four times less likely than Opus 4.7 to overlook flaws in code it generates.

Better Benchmark Scores and Faster Performance

Anthropic says Claude Opus 4.8 achieved a score of 69.2% on SWE-Bench Pro, outperforming competing models such as GPT-5.5 and Gemini 3.1 Pro on several coding and reasoning benchmarks. The company also highlighted improvements in alignment and safety testing, with lower rates of deceptive or misaligned behavior compared to Opus 4.7.

The company has also improved efficiency. Fast mode now runs at up to 2.5 times the speed and costs three times less than previous fast-mode pricing, making the model more accessible for high-volume workloads.

New Features Arrive Alongside Opus 4.8

Anthropic introduced several new features with the release. Dynamic Workflows, currently available as a research preview in Claude Code, allows the AI to plan tasks and run hundreds of parallel subagents in a single session. The feature supports large codebase migrations involving hundreds of thousands of lines of code.

The company also added Effort Control in Claude.ai and Cowork, giving users the ability to choose how much effort Claude spends on a response. In addition, developers can now update system instructions during tasks through enhancements to the Messages API.

Anthropic says it is developing lower-cost models that deliver capabilities similar to Opus while also working on a new class of AI models that exceed Opus-level intelligence. The company expects its advanced Mythos-class models to reach more customers in the coming weeks after additional safety safeguards are completed.