Skip to content

Benchmark

Anthropic Multi-Agent Search

Agentic benchmarks measure how well models carry out tasks that take several steps. Scores are percentages. Higher is better.

Models measured
2
Scores recorded
5
Sources
1
IQ weight
No weight published

Score distribution

81.2%
Mean
85.4%
Median
9.3%
Std dev
66.5%
10th percentile
92.3%
90th percentile
66.5%
Min
66.5%92.3%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Claude Opus 4.5Opus 4.5 orchestrator · Opus 4.5 subagents· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.92.3%
2Claude Opus 4.5Opus 4.5 orchestrator · Haiku 4.5 subagents· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.87%
3Claude Opus 4.5Opus 4.5 orchestrator · Sonnet 4.5 subagents· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.85.4%
4Claude Opus 4.5single-agent baseline · Opus 4.5 orchestrator· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.74.8%
5Claude Sonnet 4.5Sonnet 4.5 orchestrator · Sonnet 4.5 subagents· Reported by Anthropic· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.66.5%

Data credits: BenchBoard by SYuan03 (github.com/SYuan03/benchboard). MIT covers its code and data organization; no results licence is inferred. Numeric relays retain the original stating-source links.