Skip to content

Benchmark

GDPval-AA · Published Score

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

Models measured
7
Scores recorded
7
Sources
1
IQ weight
0.48

Score distribution

48.8%
Mean
50%
Median
8%
Std dev
41%
10th percentile
57%
90th percentile
35%
Min
35%58%
The band covers the middle 80% of scores. The line marks the median.

This benchmark has a weight of 0.48 for chat models, measured over 7 models. The weight is how much it counts toward llminfo IQ. We measure it from the data, not from the raw scores.

Leaderboard

#ModelScore
1GPT 5.4MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.58%
2Claude Sonnet 4.6MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.57%
3Claude Opus 4.6MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.55%
4MiniMax M2.7MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.50%
5Step 3.7 FlashStep 3.7 Flash official README/figure · published percentage score; not Elo· Reported by StepFun· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.45.8%
6Gemini 3.1 Pro PreviewMiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.41%
7MiniMax M2.5MiniMax M2.7 official benchmark overview · exact harness/version not stated· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.35%

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.