BBH Benchmark Leaderboard | Phaseo

BBH

AI benchmark results and model performance

Type: percentageLanguage

Recorded Results

15

Average Score

61.53%

Score Range

30.40% - 88.87%

Leading Model (lowest score)

30.40% - Ernie 4.5 21B A3B

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
Baidu	Ernie 4.5 21B A3B	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 21B A3B Thinking	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 300B A47B	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 Turbo	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 28B A3B	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
Baidu	Ernie 4.5 VL 424B A47B	-	30.40%	inferred version-family alias from ernie-4.5	Yes	Source
IBM	Granite 4.1 3B	29 Apr 2026	75.83%	3-shot, CoT	Yes	Source
Qwen	Qwen 14B	-	78.20%	inferred family alias from qwen-2.5-14b-instruct (score=0.3060; benches=16)	Yes	Source
IBM	Granite 4.1 8B	29 Apr 2026	80.51%	3-shot, CoT	Yes	Source
Qwen	Qwen 2 Math 72B	-	82.40%	inferred high-confidence family alias from qwen2-72b-instruct (score=0.4667; benches=17)	Yes	Source
Qwen	Qwen 2 Math RM 72B	-	82.40%	inferred family alias from qwen2-72b-instruct (score=0.3917; benches=17)	Yes	Source
IBM	Granite 4.1 30B	29 Apr 2026	83.74%	3-shot, CoT	Yes	Source
DeepSeek	DeepSeek V2.5 (2024-09-05)	05 Sept 2024	84.30%	inferred alias from deepseek-v2.5	Yes	Source
DeepSeek	DeepSeek V2.5 (2024-12-10)	10 Dec 2024	84.30%	inferred alias from deepseek-v2.5	Yes	Source
Qwen	Qwen 3 235B A22B	-	88.87%	-	Yes	Source