Compare AI models

Specifications
Parameters
16B/236B7B/14B
Context window
128k1M
Benchmarks
LiveCodeBench
43.4%38.6%
HumanEval
90.2%88.4%
MATH
75.7%79.5%
GSM8K
94.9%94.8%
Overview
CompanyDeepSeekQwen
Release dateJun 14 2024Jan 27 2025
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek Coder V2 leads Qwen2.5-1M on 3 of the 4 benchmarks they both report (LiveCodeBench, HumanEval, MATH, GSM8K). DeepSeek Coder V2 shipped 227 days before Qwen2.5-1M, so benchmark comparisons should account for the intervening progress.

DeepSeek Coder V2 has 16B/236B parameters, while Qwen2.5-1M has 7B/14B. Context windows are 128k (DeepSeek Coder V2) vs 1M (Qwen2.5-1M).

On LiveCodeBench, DeepSeek Coder V2 leads at 43.4% vs Qwen2.5-1M at 38.6%. On HumanEval, DeepSeek Coder V2 leads at 90.2% vs Qwen2.5-1M at 88.4%. On MATH, Qwen2.5-1M leads at 79.5% vs DeepSeek Coder V2 at 75.7%. On GSM8K, DeepSeek Coder V2 leads at 94.9% vs Qwen2.5-1M at 94.8%.