Skip to content

Benchmark

FRAMES · Agent Score

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

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

Score distribution

0.78
Mean
0.78
Median
0
Std dev
0.78
10th percentile
0.78
90th percentile
0.78
Min

Leaderboard

#ModelScore
1Mercury 2Mercury 2 official search evaluation · n=100 · medium reasoning · Exa loop · 25 calls · GPT-5.4 judge· Reported by Inception Labs· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.0.78

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.