Compare AI models

Specifications
Parameters
7B1.5B/7B/72B
Context window
4k4k
Benchmarks
MATH
51.7%85.9%
GSM8K
88.2%95.9%
Overview
CompanyDeepSeekQwen
Release dateFeb 5 2024Sep 19 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Math leads DeepSeek-Math on 2 of the 2 benchmarks they both report (MATH, GSM8K). DeepSeek-Math shipped 227 days before Qwen2.5-Math, so benchmark comparisons should account for the intervening progress.

DeepSeek-Math has 7B parameters, while Qwen2.5-Math has 1.5B/7B/72B. Context windows are 4k (DeepSeek-Math) vs 4k (Qwen2.5-Math).

On MATH, Qwen2.5-Math leads at 85.9% vs DeepSeek-Math at 51.7%. On GSM8K, Qwen2.5-Math leads at 95.9% vs DeepSeek-Math at 88.2%.