Compare AI models

Specifications
Parameters
16B—
Context window
32k1M
Benchmarks
HumanEval
57.3%86.6%
MATH
27.9%81.1%
GSM8K
72%93.8%
Overview
CompanyDeepSeekQwen
Release dateMay 16 2024Nov 15 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Turbo (1M) leads DeepSeek V2 Lite on 3 of the 3 benchmarks they both report (HumanEval, MATH, GSM8K). DeepSeek V2 Lite shipped 183 days before Qwen2.5-Turbo (1M), so benchmark comparisons should account for the intervening progress.

Context windows are 32k (DeepSeek V2 Lite) vs 1M (Qwen2.5-Turbo (1M)). DeepSeek V2 Lite is open weight, while Qwen2.5-Turbo (1M) is closed.

On HumanEval, Qwen2.5-Turbo (1M) leads at 86.6% vs DeepSeek V2 Lite at 57.3%. On MATH, Qwen2.5-Turbo (1M) leads at 81.1% vs DeepSeek V2 Lite at 27.9%. On GSM8K, Qwen2.5-Turbo (1M) leads at 93.8% vs DeepSeek V2 Lite at 72%.