Claude Sonnet 5vsGLM-5.3

Claude Sonnet 5
GLM-5.3
Specifications
Parameters
743B
Benchmarks
Nonsense detection
BullshitBench v2
80%
Prompt injection robustness
Gray Swan IPI · k = 1
0.6%
Prompt injection robustness
Gray Swan IPI · k = 10
4.7%
Prompt injection robustness
Gray Swan IPI · k = 15
5.9%
Agentic coding
SWE-Bench Pro
63.2%
Agentic coding
CursorBench v3.2
61.5%
Agentic coding
CursorBench v3.1
61.2%
Agentic coding
DeepSWE 1.1
66.9%
Next.js coding
Next.js Evals
79%
Supabase coding
Supabase Evals · with skills
95.5%
Supabase coding
Supabase Evals · no skills
90.9%
Agentic computer work
Frontier-Bench v0.1
14.6%
Agentic terminal coding
Terminal-Bench 3.0
28.3%
Agentic terminal coding
Terminal-Bench 2.1
80.4%
Web browsing
BrowseComp
84.7%
Cybersecurity
CyberGym
84.5%
Cybersecurity
ExploitBench
54.4%
Cybersecurity
ExploitGym · 6-hour budget
130
Cybersecurity
ExploitGym · 2-hour budget
105
Multidisciplinary reasoning
Humanity's Last Exam · no tools
43.2%
Multidisciplinary reasoning
Humanity's Last Exam · with tools
57.4%
62.5%Best
Agentic computer use
OSWorld-Verified
81.2%
Agentic computer use
Agent's Last Exam
28.5%
Business workflows
AutomationBench
48.2%
Knowledge work
GDPval-AA
1618
Knowledge work
GDPval-AA v2
1769
Community preference
Arena Elo (Text)
1463
Community preference (code)
Arena Elo (Code)
1543
Overview
CompanyAnthropicZ.ai
Release dateJun 30 2026Aug 14 2026
AccessProprietaryProprietary

Which is better: Claude Sonnet 5 or GLM-5.3?

GLM-5.3 leads Claude Sonnet 5 on 1 of the 1 benchmark they both report (Humanity's Last Exam). Claude Sonnet 5 shipped 45 days before GLM-5.3, so benchmark comparisons should account for the intervening progress.

Published specifications for these two models are limited — see each model page for the latest details.

On Humanity's Last Exam · with tools, GLM-5.3 leads at 62.5% vs Claude Sonnet 5 at 57.4%.

Frequently asked questions

Claude Sonnet 5 was released by Anthropic on Jun 30 2026.

GLM-5.3 was released by Z.ai on Aug 14 2026.

GLM-5.3 leads on Humanity's Last Exam · with tools — Claude Sonnet 5 57.4% vs GLM-5.3 62.5%.

Other comparisons