DeepSeek-V4-Pro-0813vsGLM-5.2

DeepSeek-V4-Pro-0813
GLM-5.2
Specifications
Parameters
744B
Context window
1M
Benchmarks
Nonsense detection
BullshitBench v2
31%
Agentic coding
SWE-Bench Pro
62.1%
Agentic coding
CursorBench v3.2
55%
Agentic coding
CursorBench v3.1
54.6%
Agentic coding
DeepSWE 1.1
44%
Next.js coding
Next.js Evals
88%
Agentic computer work
Frontier-Bench v0.1
5.1%
Agentic terminal coding
Terminal-Bench 2.1
87.9%Best
81%
Personal tool use
Toolathlon-Verified
74.1%
Cybersecurity
CyberGym
83.3%
Multidisciplinary reasoning
Humanity's Last Exam · no tools
42.7%Best
40.5%
Multidisciplinary reasoning
Humanity's Last Exam · with tools
60%Best
54.7%
Science
GPQA Diamond
91.2%
Business workflows
AutomationBench
31.8%
Knowledge work
GDPval-AA v2
1514
Community preference (code)
Arena Elo (Code)
1587
Overview
CompanyDeepSeekZ.ai
Release dateAug 13 2026Jun 16 2026
AccessProprietaryOpen Weight

Which is better: DeepSeek-V4-Pro-0813 or GLM-5.2?

DeepSeek-V4-Pro-0813 leads GLM-5.2 on 3 of the 3 benchmarks they both report (Terminal-Bench 2.1, Humanity's Last Exam). GLM-5.2 shipped 58 days before DeepSeek-V4-Pro-0813, so benchmark comparisons should account for the intervening progress.

DeepSeek-V4-Pro-0813 is proprietary, while GLM-5.2 is open weight.

On Terminal-Bench 2.1, DeepSeek-V4-Pro-0813 leads at 87.9% vs GLM-5.2 at 81%. On Humanity's Last Exam · no tools, DeepSeek-V4-Pro-0813 leads at 42.7% vs GLM-5.2 at 40.5%. On Humanity's Last Exam · with tools, DeepSeek-V4-Pro-0813 leads at 60% vs GLM-5.2 at 54.7%.

Frequently asked questions

DeepSeek-V4-Pro-0813 was released by DeepSeek on Aug 13 2026.

GLM-5.2 was released by Z.ai on Jun 16 2026.

DeepSeek-V4-Pro-0813 leads on Terminal-Bench 2.1 — DeepSeek-V4-Pro-0813 87.9% vs GLM-5.2 81%.

DeepSeek-V4-Pro-0813 leads on Humanity's Last Exam · no tools — DeepSeek-V4-Pro-0813 42.7% vs GLM-5.2 40.5%.

DeepSeek-V4-Pro-0813 is a proprietary model released by DeepSeek. GLM-5.2 is an open weight model released by Z.ai.

Other comparisons