BrowseComp-zh Benchmark Leaderboard | Phaseo

BrowseComp-zh

AI benchmark results and model performance

Type: percentageReasoning

Recorded Results

15

Average Score

61.23%

Score Range

47.90% - 82.40%

Leading Model (lowest score)

47.90% - DeepSeek OCR 2

Scores Over Time

Individual benchmark scores plotted by date.

Models Using This Benchmark

Organisation	Model	Reported	Top Score	Info	Self Reported	Source
DeepSeek	DeepSeek OCR 2	-	47.90%	inferred family alias from deepseek-v3.2-exp (score=0.3809; benches=14)	Yes	Source
DeepSeek	DeepSeek V3.2 Exp	29 Sept 2025	47.90%	-	Yes	Source
MiniMax	MiniMax M2	27 Oct 2025	48.50%	-	Yes	-
MiniMax	MiniMax M2 Her	24 Jan 2026	48.50%	inferred modality/version alias from minimax-m2	Yes	-
DeepSeek	DeepSeek V3.1	21 Aug 2025	49.20%	Search agent with commercial API + webpage filter + 128K context	Yes	Source
DeepSeek	DeepSeek V3.1 Terminus	22 Sept 2025	49.20%	inferred alias from deepseek-v3.1	Yes	Source
Qwen	Qwen 3.5 27B	24 Feb 2026	62.10%	-	Yes	Source
Qwen	Qwen 3.5 Flash	23 Feb 2026	62.10%	inferred family alias from qwen3.5-27b (score=0.4147; benches=81)	Yes	Source
Moonshot	Kimi K2 Thinking	06 Nov 2025	62.30%	inferred alias from kimi-k2-thinking-0905	Yes	Source
z.AI	GLM 4.7	22 Dec 2025	66.60%	-	Yes	Source
Qwen	Qwen 3.5 35B A3B	24 Feb 2026	69.50%	-	Yes	Source
Qwen	Qwen 3.5 122B A10B	24 Feb 2026	69.90%	-	Yes	Source
Qwen	Qwen 3.5 397B A17B	16 Feb 2026	70.30%	-	Yes	Source
ByteDance	Seed 2.0 Lite	14 Feb 2026	82%	Seed2 official benchmark table \| BrowseComp-zh	Yes	Source
ByteDance	Seed 2.0 Pro	14 Feb 2026	82.40%	Seed2 official benchmark table \| BrowseComp-zh	Yes	Source