Claude 4 Opus

MODEL RECORD: ANTHR-O4

Organization
Anthropic
Family
Claude
Model number
4
Tier
O
Status
ACTIVE
Release date
2025-05-22

Table may be scrolled horizontally on narrow displays.

Record No. Model Benchmark Metric Score Eval Date Source Model ID
BR-002O4-001 Claude 4 Opus Artificial Analysis Intelligence Index Overall 25.5 Unknown Primary source ANTHR-O4
BR-002O4-002 Claude 4 Opus GPQA Diamond Accuracy 70.1% Unknown Primary source ANTHR-O4
BR-002O4-003 Claude 4 Opus Humanity's Last Exam No tools Overall 5.9% Unknown Primary source ANTHR-O4
BR-002O4-004 Claude 4 Opus SciCode Overall 40.9% Unknown Primary source ANTHR-O4
BR-002O4-005 Claude 4 Opus IFBench Overall 43.3% Unknown Primary source ANTHR-O4
BR-002O4-006 Claude 4 Opus AA-LCR Overall 36% Unknown Primary source ANTHR-O4
BR-002O4-007 Claude 4 Opus Artificial Analysis Intelligence Index Overall 31 Unknown Primary source ANTHR-O4
BR-002O4-008 Claude 4 Opus GPQA Diamond Accuracy 79.6% Unknown Primary source ANTHR-O4
BR-002O4-009 Claude 4 Opus Humanity's Last Exam No tools Overall 11.7% Unknown Primary source ANTHR-O4
BR-002O4-010 Claude 4 Opus SciCode Overall 39.8% Unknown Primary source ANTHR-O4
BR-002O4-011 Claude 4 Opus tau2-bench Overall 73.4% Unknown Primary source ANTHR-O4
BR-002O4-012 Claude 4 Opus Terminal-Bench Hard Overall 31.1% Unknown Primary source ANTHR-O4
BR-002O4-013 Claude 4 Opus IFBench Overall 53.7% Unknown Primary source ANTHR-O4
BR-002O4-014 Claude 4 Opus AA-LCR Overall 33.7% Unknown Primary source ANTHR-O4