Compare AI models

Specifications
Parameters
671B110B
Context window
128k32k
Benchmarks
MMLU
90.8%80.4%
Overview
CompanyDeepSeekQwen
Release dateJan 20 2025Apr 25 2024
AccessOpen WeightOpen Weight
Model detailsView modelView model

Frequently asked questions

DeepSeek R1 leads Qwen1.5-110B on 1 of the 1 benchmark they both report (MMLU). Qwen1.5-110B shipped 270 days before DeepSeek R1, so benchmark comparisons should account for the intervening progress.

DeepSeek R1 has 671B parameters, while Qwen1.5-110B has 110B. Context windows are 128k (DeepSeek R1) vs 32k (Qwen1.5-110B).

On MMLU, DeepSeek R1 leads at 90.8% vs Qwen1.5-110B at 80.4%.