Compare AI models

Specifications
Parameters
—6B
Context window
128k8k
Benchmarks
HumanEval
83.7%58.5%
MATH
64.3%25.7%
MMLU
84.7%61.4%
Overview
CompanyOpenAIZ.ai
Release dateNov 6 2023Oct 27 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

GPT-4 Turbo leads ChatGLM3-6B on 3 of the 3 benchmarks they both report (HumanEval, MATH, MMLU). ChatGLM3-6B shipped 10 days before GPT-4 Turbo, so benchmark comparisons should account for the intervening progress.

Context windows are 128k (GPT-4 Turbo) vs 8k (ChatGLM3-6B). GPT-4 Turbo is closed, while ChatGLM3-6B is open weight.

On HumanEval, GPT-4 Turbo leads at 83.7% vs ChatGLM3-6B at 58.5%. On MATH, GPT-4 Turbo leads at 64.3% vs ChatGLM3-6B at 25.7%. On MMLU, GPT-4 Turbo leads at 84.7% vs ChatGLM3-6B at 61.4%.