gpt.buzz
Sign in

Compare models

Pick up to 4models. Specs render side-by-side. Share the URL — it's stateless.

SelectedDeepSeek logoDeepSeek-V3.1×DeepSeek logoDeepSeek-R1×Mistral logoMistral Large 2×OpenAI logoo3×Clear all
 
DeepSeek logoDeepSeek-V3.1

DeepSeek

DeepSeek logoDeepSeek-R1

DeepSeek

Mistral logoMistral Large 2

Mistral

OpenAI logoo3

OpenAI

VendorDeepSeekDeepSeekMistralOpenAI
FamilyDeepSeekDeepSeekMistralo-series
Release date2025-08-212025-01-202024-07-242025-04-16
Context window128,000 tokens128,000 tokens128,000 tokens200,000 tokens
Parameters671B671B123B
Modalitytexttexttexttext, vision
LicenseMITMITMistral Research Licenseproprietary
Sourceopen weightsopen weightsopen weightsproprietary
DescriptionLarge MoE open-weight model. Predecessor to DeepSeek-V4.Reasoning-focused open-weight model.Reasoning-focused model in the o-series.
Links

Add a model

Max 4 models. Remove one to add another.