Neura Intelligence Index

AI Model Benchmarks

Back to Leaderboard

DeepSeek-V4-Pro-Max vs GLM-5.2

DeepSeek-V4-Pro-Max

DeepSeek 🇨🇳

73.1

#64—

Z

Zhipu AI 🇨🇳

74.6

#61—

Category Comparison

Benchmark-by-Benchmark

reasoning

Benchmark	DeepSeek-V4-Pro-Max	GLM-5.2
GPQA Diamond	90.1	91.2
Humanity's Last Exam	48.2	54.7

coding

Benchmark	DeepSeek-V4-Pro-Max	GLM-5.2
SWE-bench Verified	80.6	—
SWE-Bench Pro	55.4	62.1

general

Benchmark	DeepSeek-V4-Pro-Max	GLM-5.2
SimpleQA	57.9	—

agent

Benchmark	DeepSeek-V4-Pro-Max	GLM-5.2
BrowseComp	83.4	—
Toolathlon	51.8	48.2
MCP Atlas	73.6	76.8

Category Winners

coding

Z

GLM-5.2

65.8

reasoning

Z

GLM-5.2

88.7

general

DeepSeek-V4-Pro-Max

78.1

agent

DeepSeek-V4-Pro-Max

62.8

DeepSeek-V4-Pro-Max vs GLM-5.2 — Benchmark Comparison | Neura Market