Timeline
Claude Opus 4.8 beats Gemini Pro 5 by 11 points on Fable 5 benchmark
Anthropic releases Claude Opus 4.7 with 92% honesty benchmark and reduced sycophancy
Claude Opus 4.8 achieves 89% task completion and 2.5% harm rate on WorkBench, a dramatic improvement over GPT-4.
Claude Opus 4.8 adds dynamic workflows for agentic coding
Claude Opus 4.8 launched with dynamic workflows for Claude Code, enabling multi-step agentic coding.
Used as CEO agent in 11-agent experiment that earned $0 revenue
Autonomously ported Adobe Lightroom CC to Linux via Wine after a single prompt
Claude market share reached 10.3% with 13% subscription conversion rate.
Opus 4.7 released with new tokenizer causing 40%+ cost increase
Anthropic released Claude Opus 4.7 with an updated tokenizer that increases token counts for the same input
Ecosystem
Claude Opus 4.7
Claude Opus 4.6
Benchmarks
Evidence (15 articles)
Anthropic's Opus 4.7 Shows Sustained Gains on Economically Critical Tasks
Apr 18, 2026Claude Opus 4.7 Appears on Anthropic's Internal API, Hinting at Imminent Release
Apr 12, 2026Claude 'Mythos' Leak Suggests New Tier Beyond Opus 4.6, Targeting Cybersecurity Partners First
Mar 27, 2026Anthropic's Claude Sonnet 4.8, Opus 4.7 Internally Tested, Leak Suggests
Apr 6, 2026Anthropic Deprecates Fixed Thinking Budgets, Forces Adaptive Mode
May 14, 2026Anthropic Ships Claude Opus 4.7: 80.1 SWE-Bench, 1M Context
May 17, 2026Cursor Announces Composer 2: Smaller, Cheaper Coding-Specific Model Targeting Claude Opus Performance
Mar 19, 2026Anthropic Claims Claude Opus 4.7 Hits 92% Honesty, Cuts Sycophancy
Jul 6, 2026+ 7 more articles