Claude Opus 4.8 (Non-reasoning): benchmark results

Provider: Anthropic. Released 2026-05-28. Access: API.

Aggregate Rating 1708 ± 22, rank #362 of 2286 rated models, from 30 benchmark results.

Strongest benchmark results

BenchmarkScoreMetricPercentile
German LLM Index - ScaLA Acceptability82.62Matthews correlation coefficient (x100)95.3
German LLM Index - MuSR (564 questions)86.35Accuracy (%)86.1
German LLM Index - SB10K Sentiment64.52Matthews correlation coefficient (x100)86
StudentBench - Practice Design0.96Expert tutor pairwise preference, Bradley-Terry score (pract83.3
WeirdML70.45Average Score81.4
SteerBench-Work - Pass^584.9Scenarios Correct in All 5 Trials (%)79.3
German LLM Index - GermEval NER85.84seqeval micro-F1 excluding MISC (%)77.3
German LLM Index - INCLUDE (139 questions)71.94Exact match (%)70.4
K-Bench (Therapeutic v0)97.81Overall (%, Therapeutic v0 prompt)67.7
ComboShoppingBench - Coupon Legality95.9Pass rate (%)66.7
StudentBench - Lesson Planning0.59Expert tutor pairwise preference, Bradley-Terry score (lesso66.7
K-Bench98.11Overall (%, Default v1 prompt)65.6

Interactive version: theaggregate.ai/model?slug=claude-opus-4-8-non-reasoning · How It Works · Data refreshed daily, snapshot 2026-11-20.