Compare AI models

Specifications
Parameters
7B110B
Context window
4k32k
Benchmarks
MATH
51.7%49.6%
GSM8K
88.2%85.4%
Overview
CompanyDeepSeekQwen
Release dateFeb 5 2024Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek-Math leads Qwen1.5-110B on 2 of the 2 benchmarks they both report (MATH, GSM8K). DeepSeek-Math shipped 80 days before Qwen1.5-110B, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while Qwen1.5-110B has 110B. Context windows are 4k (DeepSeek-Math) vs 32k (Qwen1.5-110B).

On MATH, DeepSeek-Math leads at 51.7% vs Qwen1.5-110B at 49.6%. On GSM8K, DeepSeek-Math leads at 88.2% vs Qwen1.5-110B at 85.4%.