Grok 4.5vsGLM-5.2

Grok 4.5
GLM-5.2
Specifications
Parameters
744B
Context window
1M
Benchmarks
Nonsense detection
BullshitBench v2
54%Best
31%
Prompt injection robustness
Gray Swan IPI · k = 1
13.4%
Prompt injection robustness
Gray Swan IPI · k = 10
54.2%
Prompt injection robustness
Gray Swan IPI · k = 15
60.8%
Agentic coding
SWE-Bench Pro
64.7%Best
62.1%
Multilingual coding
SWE-Bench Multilingual
78%
Agentic coding
CursorBench v3.2
66.7%Best
55%
Agentic coding
CursorBench v3.1
54.6%
Agentic coding
DeepSWE 1.1
54%Best
44%
Agentic coding
DeepSWE 1.0
62%
Agentic coding
FrontierCode v1.1 (Extended) · extended split
56.6%
Expert software engineering
APEX-SWE
53.6%
Next.js coding
Next.js Evals
83%
88%Best
Agentic computer work
Frontier-Bench v0.1
17.8%Best
5.1%
Agentic terminal coding
Terminal-Bench 3.0
15.7%
Agentic terminal coding
Terminal-Bench 2.1
83.3%Best
81%
Expert agentic work
APEX-Agents
47.1%
Multidisciplinary reasoning
Humanity's Last Exam · no tools
40.5%
Multidisciplinary reasoning
Humanity's Last Exam · with tools
54.7%
Science
GPQA Diamond
91.2%
Agentic legal work
Harvey's Legal Agent Benchmark
12.92%
Medical admin work
MedScribe
86.88%
Overall intelligence
AA Intelligence Index
56
Knowledge work
GDPval-AA v2
1526Best
1514
Knowledge work
AA-Briefcase
1313
Community preference
Arena Elo (Text)
1468
Community preference (code)
Arena Elo (Code)
1555
1587Best
Overview
CompanySpaceXAIZ.ai
Release dateJul 8 2026Jun 16 2026
AccessProprietaryOpen Weight

Which is better: Grok 4.5 or GLM-5.2?

Grok 4.5 leads GLM-5.2 on 7 of the 9 benchmarks they both report. GLM-5.2 shipped 22 days before Grok 4.5, so benchmark comparisons should account for the intervening progress.

Grok 4.5 is proprietary, while GLM-5.2 is open weight.

On BullshitBench v2, Grok 4.5 leads at 54% vs GLM-5.2 at 31%. On SWE-Bench Pro, Grok 4.5 leads at 64.7% vs GLM-5.2 at 62.1%. On CursorBench v3.2, Grok 4.5 leads at 66.7% vs GLM-5.2 at 55%. On DeepSWE 1.1, Grok 4.5 leads at 54% vs GLM-5.2 at 44%. On Next.js Evals, GLM-5.2 leads at 88% vs Grok 4.5 at 83%. On Frontier-Bench v0.1, Grok 4.5 leads at 17.8% vs GLM-5.2 at 5.1%. On Terminal-Bench 2.1, Grok 4.5 leads at 83.3% vs GLM-5.2 at 81%. On GDPval-AA v2, Grok 4.5 leads at 1526 vs GLM-5.2 at 1514. On Arena Elo (Code), GLM-5.2 leads at 1587 vs Grok 4.5 at 1555.

Frequently asked questions

Grok 4.5 was released by SpaceXAI on Jul 8 2026.

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

Grok 4.5 leads on SWE-Bench Pro — Grok 4.5 64.7% vs GLM-5.2 62.1%.

Grok 4.5 is a proprietary model released by SpaceXAI. GLM-5.2 is an open weight model released by Z.ai.

Other comparisons