Compare AI models

Specifications
Parameters
236B—
Context window
128k1M
Benchmarks
HumanEval
84.76%86.6%
MATH
71.02%81.1%
Overview
CompanyDeepSeekQwen
Release dateJun 28 2024Nov 15 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Turbo (1M) leads DeepSeek V2-0628 on 2 of the 2 benchmarks they both report (HumanEval, MATH). DeepSeek V2-0628 shipped 140 days before Qwen2.5-Turbo (1M), so benchmark comparisons should account for the intervening progress.

Context windows are 128k (DeepSeek V2-0628) vs 1M (Qwen2.5-Turbo (1M)). DeepSeek V2-0628 is open weight, while Qwen2.5-Turbo (1M) is closed.

On HumanEval, Qwen2.5-Turbo (1M) leads at 86.6% vs DeepSeek V2-0628 at 84.76%. On MATH, Qwen2.5-Turbo (1M) leads at 81.1% vs DeepSeek V2-0628 at 71.02%.