gpt-oss-120bvsGLM-5.3

gpt-oss-120b
GLM-5.3
Specifications
Parameters
117B
743B
Context window
128k
Benchmarks
Nonsense detection
BullshitBench v2
11%
Coding
SWE-Bench Verified
62.4%
Agentic coding
DeepSWE 1.1
66.9%
Agentic terminal coding
Terminal-Bench 3.0
28.3%
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
14.9%
Multidisciplinary reasoning
Humanity's Last Exam · with tools
19%
62.5%Best
Science
GPQA Diamond
80.1%
General knowledge
MMLU
90%
Agentic computer use
Agent's Last Exam
28.5%
Business workflows
AutomationBench
48.2%
Knowledge work
GDPval-AA v2
1769
Overview
CompanyOpenAIZ.ai
Release dateAug 5 2025Aug 14 2026
AccessOpen WeightProprietary

Which is better: gpt-oss-120b or GLM-5.3?

GLM-5.3 leads gpt-oss-120b on 1 of the 1 benchmark they both report (Humanity's Last Exam). gpt-oss-120b shipped 374 days before GLM-5.3, so benchmark comparisons should account for the intervening progress.

gpt-oss-120b has 117B parameters, while GLM-5.3 has 743B. gpt-oss-120b is open weight, while GLM-5.3 is proprietary.

On Humanity's Last Exam · with tools, GLM-5.3 leads at 62.5% vs gpt-oss-120b at 19%.

Frequently asked questions

gpt-oss-120b was released by OpenAI on Aug 5 2025.

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

GLM-5.3 leads on Humanity's Last Exam · with tools — gpt-oss-120b 19% vs GLM-5.3 62.5%.

gpt-oss-120b is an open weight model released by OpenAI. GLM-5.3 is a proprietary model released by Z.ai.

Other comparisons