Skip to content

Benchmark

GDPval-MM

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

Models measured
3
Scores recorded
6
Sources
2
IQ weight
No weight published

Score distribution

67.2%
Mean
59%
Median
11.6%
Std dev
59%
10th percentile
84.9%
90th percentile
59%
Min
59%84.9%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1GPT 5.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim84.9%
2GPT-5.5 Pro· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim82.3%
3MiniMax M2.5· LLM Stats data· Reported by the model's developer. The source relaying it did not verify it. claim59%
4MiniMax M2.5main capability figure· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59%
5MiniMax M2.5office figure· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59%
6MiniMax M2.5official prose · average win rate· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.59%

Data credits: Data by LLM Stats · 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.