Compare AI models

Specifications
Parameters
7B/67B1.5B/7B/72B
Context window
4k4k
Benchmarks
MATH
32.6%85.9%
GSM8K
84.1%95.9%
Overview
CompanyDeepSeekQwen
Release dateNov 29 2023Sep 19 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Math leads DeepSeek-LLM on 2 of the 2 benchmarks they both report (MATH, GSM8K). DeepSeek-LLM shipped 295 days before Qwen2.5-Math, so benchmark comparisons should account for the intervening progress.

DeepSeek-LLM has 7B/67B parameters, while Qwen2.5-Math has 1.5B/7B/72B. Context windows are 4k (DeepSeek-LLM) vs 4k (Qwen2.5-Math).

On MATH, Qwen2.5-Math leads at 85.9% vs DeepSeek-LLM at 32.6%. On GSM8K, Qwen2.5-Math leads at 95.9% vs DeepSeek-LLM at 84.1%.