GPT-5.5-ProvsGLM-5.3

GPT-5.5-Pro
GLM-5.3
Specifications
Parameters
743B
Benchmarks
Nonsense detection
BullshitBench v2
36%
Agentic coding
DeepSWE 1.1
66.9%
Next.js coding
Next.js Evals
83%
Agentic terminal coding
Terminal-Bench 3.0
28.3%
Web browsing
BrowseComp
90.1%
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 · with tools
57.2%
62.5%Best
Advanced math
FrontierMath · Tier 1–3
52.4%
Advanced math
FrontierMath · Tier 4
39.6%
Agentic computer use
Agent's Last Exam
28.5%
Business workflows
AutomationBench
48.2%
Knowledge work
GDPval-AA v2
1769
Knowledge work
GDPval (win/tie rate)
82.3%
Overview
CompanyOpenAIZ.ai
Release dateApr 23 2026Aug 14 2026
AccessProprietaryProprietary

Which is better: GPT-5.5-Pro or GLM-5.3?

GLM-5.3 leads GPT-5.5-Pro on 1 of the 1 benchmark they both report (Humanity's Last Exam). GPT-5.5-Pro shipped 113 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 GPT-5.5-Pro at 57.2%.

Frequently asked questions

GPT-5.5-Pro was released by OpenAI on Apr 23 2026.

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

GLM-5.3 leads on Humanity's Last Exam · with tools — GPT-5.5-Pro 57.2% vs GLM-5.3 62.5%.

Other comparisons