Compare AI models

Specifications
Parameters
16B110B
Context window
32k32k
Benchmarks
HumanEval
57.3%52.4%
MATH
27.9%49.6%
MMLU
55.7%80.4%
GSM8K
72%85.4%
Overview
CompanyDeepSeekQwen
Release dateMay 16 2024Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen1.5-110B leads DeepSeek V2 Lite on 3 of the 4 benchmarks they both report (HumanEval, MATH, MMLU, GSM8K). Qwen1.5-110B shipped 21 days before DeepSeek V2 Lite, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 Lite has 16B parameters, while Qwen1.5-110B has 110B. Context windows are 32k (DeepSeek V2 Lite) vs 32k (Qwen1.5-110B).

On HumanEval, DeepSeek V2 Lite leads at 57.3% vs Qwen1.5-110B at 52.4%. On MATH, Qwen1.5-110B leads at 49.6% vs DeepSeek V2 Lite at 27.9%. On MMLU, Qwen1.5-110B leads at 80.4% vs DeepSeek V2 Lite at 55.7%. On GSM8K, Qwen1.5-110B leads at 85.4% vs DeepSeek V2 Lite at 72%.