Compare AI models

Specifications
Context window
1M128k
Benchmarks
HumanEval
86.6%75.7%
IFEval
76.3%75.2%
MATH
81.1%57.9%
GSM8K
93.8%90.9%
Overview
CompanyQwenZ.ai
Release dateNov 15 2024Jun 5 2024
AccessClosedClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Turbo (1M) leads GLM-4-Air on 4 of the 4 benchmarks they both report (HumanEval, IFEval, MATH, GSM8K). GLM-4-Air shipped 163 days before Qwen2.5-Turbo (1M), so benchmark comparisons should account for the intervening progress.

Context windows are 1M (Qwen2.5-Turbo (1M)) vs 128k (GLM-4-Air).

On HumanEval, Qwen2.5-Turbo (1M) leads at 86.6% vs GLM-4-Air at 75.7%. On IFEval, Qwen2.5-Turbo (1M) leads at 76.3% vs GLM-4-Air at 75.2%. On MATH, Qwen2.5-Turbo (1M) leads at 81.1% vs GLM-4-Air at 57.9%. On GSM8K, Qwen2.5-Turbo (1M) leads at 93.8% vs GLM-4-Air at 90.9%.