Compare AI models

Specifications
Parameters
7B/67B—
Context window
4k1M
Benchmarks
HumanEval
73.8%86.6%
MATH
32.6%81.1%
GSM8K
84.1%93.8%
Overview
CompanyDeepSeekQwen
Release dateNov 29 2023Nov 15 2024
AccessOpen WeightClosed
Model detailsView modelView model

Frequently asked questions

Qwen2.5-Turbo (1M) leads DeepSeek-LLM on 3 of the 3 benchmarks they both report (HumanEval, MATH, GSM8K). DeepSeek-LLM shipped 352 days before Qwen2.5-Turbo (1M), so benchmark comparisons should account for the intervening progress.

Context windows are 4k (DeepSeek-LLM) vs 1M (Qwen2.5-Turbo (1M)). DeepSeek-LLM is open weight, while Qwen2.5-Turbo (1M) is closed.

On HumanEval, Qwen2.5-Turbo (1M) leads at 86.6% vs DeepSeek-LLM at 73.8%. On MATH, Qwen2.5-Turbo (1M) leads at 81.1% vs DeepSeek-LLM at 32.6%. On GSM8K, Qwen2.5-Turbo (1M) leads at 93.8% vs DeepSeek-LLM at 84.1%.