Compare AI models

Specifications
Parameters
—6B
Context window
100k32k
Benchmarks
HumanEval
71.2%9.8%
MMLU
78.5%45.2%
GSM8K
88%25.9%
Overview
CompanyAnthropicZ.ai
Release dateJul 11 2023Jun 25 2023
AccessClosedOpen Weight
Model detailsView modelView model

Frequently asked questions

Claude 2 leads ChatGLM2-6B on 3 of the 3 benchmarks they both report (HumanEval, MMLU, GSM8K). ChatGLM2-6B shipped 16 days before Claude 2, so benchmark comparisons should account for the intervening progress.

Context windows are 100k (Claude 2) vs 32k (ChatGLM2-6B). Claude 2 is closed, while ChatGLM2-6B is open weight.

On HumanEval, Claude 2 leads at 71.2% vs ChatGLM2-6B at 9.8%. On MMLU, Claude 2 leads at 78.5% vs ChatGLM2-6B at 45.2%. On GSM8K, Claude 2 leads at 88% vs ChatGLM2-6B at 25.9%.