Compare AI models

Specifications
Parameters
236B7B/14B
Context window
128k1M
Benchmarks
LiveCodeBench
32.5%38.6%
HumanEval
81.1%88.4%
MATH
53.9%79.5%
GSM8K
92.2%94.8%
Overview
CompanyDeepSeekQwen
Release dateMay 6 2024Jan 27 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

Qwen2.5-1M leads DeepSeek V2 on 4 of the 4 benchmarks they both report (LiveCodeBench, HumanEval, MATH, GSM8K). DeepSeek V2 shipped 266 days before Qwen2.5-1M, so benchmark comparisons should account for the intervening progress.

DeepSeek V2 has 236B parameters, while Qwen2.5-1M has 7B/14B. Context windows are 128k (DeepSeek V2) vs 1M (Qwen2.5-1M).

On LiveCodeBench, Qwen2.5-1M leads at 38.6% vs DeepSeek V2 at 32.5%. On HumanEval, Qwen2.5-1M leads at 88.4% vs DeepSeek V2 at 81.1%. On MATH, Qwen2.5-1M leads at 79.5% vs DeepSeek V2 at 53.9%. On GSM8K, Qwen2.5-1M leads at 94.8% vs DeepSeek V2 at 92.2%.