Compare AI models

Specifications
Parameters
110B6B
Context window
32k2k
Benchmarks
HumanEval
52.4%0%
MATH
49.6%3.1%
HellaSwag
87.5%30.4%
MMLU
80.4%25.2%
GSM8K
85.4%1.5%
Overview
CompanyQwenZ.ai
Release dateApr 25 2024Mar 14 2023
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-110B leads ChatGLM-6B on 5 of the 5 benchmarks they both report. ChatGLM-6B shipped 408 days before Qwen1.5-110B, so benchmark comparisons should account for the intervening progress.

Qwen1.5-110B has 110B parameters, while ChatGLM-6B has 6B. Context windows are 32k (Qwen1.5-110B) vs 2k (ChatGLM-6B).

On HumanEval, Qwen1.5-110B leads at 52.4% vs ChatGLM-6B at 0%. On MATH, Qwen1.5-110B leads at 49.6% vs ChatGLM-6B at 3.1%. On HellaSwag, Qwen1.5-110B leads at 87.5% vs ChatGLM-6B at 30.4%. On MMLU, Qwen1.5-110B leads at 80.4% vs ChatGLM-6B at 25.2%. On GSM8K, Qwen1.5-110B leads at 85.4% vs ChatGLM-6B at 1.5%.