Compare AI models

Specifications
Parameters
70B110B
Context window
16k32k
Benchmarks
HumanEval
53%52.4%
Overview
CompanyMetaQwen
Release dateJan 29 2024Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Code Llama 70B leads Qwen1.5-110B on 1 of the 1 benchmark they both report (HumanEval). Code Llama 70B shipped 87 days before Qwen1.5-110B, so benchmark comparisons should account for the intervening progress.

Code Llama 70B has 70B parameters, while Qwen1.5-110B has 110B. Context windows are 16k (Code Llama 70B) vs 32k (Qwen1.5-110B).

On HumanEval, Code Llama 70B leads at 53% vs Qwen1.5-110B at 52.4%.