Compare AI models

Specifications
Context window
128k128k
Benchmarks
HumanEval
74.1%75.7%
MATH
50.6%57.9%
MMLU
81.3%81.9%
GSM8K
90%90.9%
Overview
CompanySpaceXSIZ.ai
Release dateMar 28 2024Jun 5 2024
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

GLM-4-Air leads Grok‑1.5 on 4 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Grok‑1.5 shipped 69 days before GLM-4-Air, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (Grok‑1.5) vs 128k (GLM-4-Air).

On HumanEval, GLM-4-Air leads at 75.7% vs Grok‑1.5 at 74.1%. On MATH, GLM-4-Air leads at 57.9% vs Grok‑1.5 at 50.6%. On MMLU, GLM-4-Air leads at 81.9% vs Grok‑1.5 at 81.3%. On GSM8K, GLM-4-Air leads at 90.9% vs Grok‑1.5 at 90%.