Composer 2.5vsGLM-5

Composer 2.5
GLM-5
Specifications
Parameters
744B
Benchmarks
Nonsense detection
BullshitBench v2
28%
Agentic coding
SWE-Bench Pro
54%
Coding
SWE-Bench Verified
77.8%
Multilingual coding
SWE-Bench Multilingual
79.8%Best
73.3%
Agentic coding
CursorBench v3.2
56.1%
Agentic coding
CursorBench v3.1
63.2%
Agentic coding
DeepSWE 1.0
18%
Next.js coding
Next.js Evals
92%
Agentic terminal coding
Terminal-Bench 2.1
73%
Agentic terminal coding
Terminal-Bench 2.0
69.3%Best
56.2%
Web browsing
BrowseComp
75.9%
Multidisciplinary reasoning
Humanity's Last Exam · with tools
50.4%
Science
GPQA Diamond
86%
Community preference (code)
Arena Elo (Code)
1435
Overview
CompanyCursorZ.ai
Release dateMay 18 2026Feb 12 2026
AccessProprietaryOpen Weight

Which is better: Composer 2.5 or GLM-5?

Composer 2.5 leads GLM-5 on 2 of the 2 benchmarks they both report (SWE-Bench Multilingual, Terminal-Bench 2.0). GLM-5 shipped 95 days before Composer 2.5, so benchmark comparisons should account for the intervening progress.

Composer 2.5 is proprietary, while GLM-5 is open weight.

On SWE-Bench Multilingual, Composer 2.5 leads at 79.8% vs GLM-5 at 73.3%. On Terminal-Bench 2.0, Composer 2.5 leads at 69.3% vs GLM-5 at 56.2%.

Frequently asked questions

Composer 2.5 was released by Cursor on May 18 2026.

GLM-5 was released by Z.ai on Feb 12 2026.

Composer 2.5 leads on SWE-Bench Multilingual — Composer 2.5 79.8% vs GLM-5 73.3%.

Composer 2.5 is a proprietary model released by Cursor. GLM-5 is an open weight model released by Z.ai.

Other comparisons