Skip to content

Benchmark

Agentic IF Index

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

Models measured
4
Scores recorded
4
Sources
1
IQ weight
No weight published

Score distribution

55.9
Mean
59.1
Median
5.7
Std dev
46.2
10th percentile
60.5
90th percentile
46.2
Min
46.260.5
The band covers the middle 80% of scores. The line marks the median.

Results

The source marks these results as descriptive, not capability scores, so we list them by name instead of ranking them.

ModelScore
Claude Opus 5max · Meta internal composite · agentic instruction following· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59.1
GPT-5.6 Solmax · Meta internal composite · agentic instruction following· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.60.5
Muse Spark 1.2xhigh · Meta internal composite · agentic instruction following· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.46.2
Muse Spark 1.3max · Meta internal composite · agentic instruction following· Reported by Meta· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.57.8

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.