Introducing Claude Opus 4.8
- Source
- anthropic.com
- Author
- Anthropic News
- Date

Context
The Opus 4.8 launch gave Claude users more control over how much effort a response receives. In its May 28, 2026 announcement, Anthropic introduced an effort control in claude.ai: higher settings make the model think more frequently and deeply, while lower settings favor faster responses and use up rate limits more slowly.
This makes the quality-versus-waiting-time tradeoff an explicit choice within the same model. A user can allocate more effort to difficult work and less to a simpler request. The setting is separate from the launch's fast mode, which Anthropic described as working at 2.5 times the speed and being three times cheaper than fast mode for previous models.
“We’re upgrading Claude Opus to a new version: Claude Opus 4.8. It builds on Opus 4.7 with improvements across benchmarks, and is a more effective collaborator. It’s available today for the same price.”
“And fast mode for Opus 4.8—where the model can work at 2.5× the speed—is now three times cheaper than it was for previous models.”
“In Claude Code, it asks the right questions, catches its own mistakes, pushes back when a plan isn’t sound, and builds up confidence around complex, multi-service explorations before making big changes.”
Tom Pritchard
“On our Super-Agent benchmark, Claude Opus 4.8 is the only model to complete every case end-to-end, beating prior Opus models and GPT-5.5 at parity on cost.”
Kay Zhu
“Claude Opus 4.8 is the strongest computer-use and browser-agent model we’ve tested, scoring 84% on Online-Mind2Web, which is a meaningful jump over both Opus 4.7 and GPT-5.5.”
Miguel Gonzalez
Checking sign-in…
Loading comments…





