Skip to content

Benchmark

Terminal-Bench 2 · Version Unspecified

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

Models measured
3
Scores recorded
3
Sources
1
IQ weight
No weight published

Score distribution

45.6%
Mean
51.7%
Median
9.2%
Std dev
32.6%
10th percentile
52.5%
90th percentile
32.6%
Min
32.6%52.5%
The band covers the middle 80% of scores. The line marks the median.

Leaderboard

#ModelScore
1Qwen3.5 397B A17BQwen3.5 release/model card · published setting· Reported by Qwen· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.52.5%
2MiniMax M2.5main capability figure · Terminal Bench 2· Reported by MiniMax· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.51.7%
3Devstral 2 123b Instruct 2512official model card· Reported by Mistral AI· via BenchBoard docs· Reported by the model's developer. The source relaying it did not verify it.32.6%

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.