Anthropic's Claude Opus 4.8 boosts agentic reasoning, cuts fast mode cost 3x

Anthropic has launched Claude Opus 4.8, the latest iteration of its flagship model, available immediately at the same price as Opus 4.7. The upgrade delivers notable improvements across agentic reasoning, coding, and knowledge work benchmarks, while introducing new features like user-controlled effort levels on claude.ai and dynamic workflows in Claude Code for tackling large-scale problems. A standout change is the price drop for Opus 4.8's fast mode—now 2.5× faster and three times cheaper than its predecessor, making high-speed inference more accessible for production workloads.

Early tester feedback underscores the model's reliability and judgment. On the Super-Agent benchmark, Opus 4.8 became the only model to complete every case end-to-end, outperforming GPT-5.5 at similar cost. It scored 84% on Online-Mind2Web for browser agents, a meaningful jump over both Opus 4.7 and GPT-5.5. In legal applications, it achieved the highest Legal Agent Benchmark score ever, becoming the first model to exceed 10% on the all-pass standard—a threshold that directly translates to more attorney work that can be confidently automated. Enterprises using Databricks' Genie agent reported a 61% reduction in token costs compared to Opus 4.7, while Hebbia and CoCounsel highlighted improvements in consistency and proactive error detection. Developers building on Devin and using Claude Code noted cleaner tool calling and higher-quality analysis outputs.