Math Misconception Test

Deepseek R1's response to the Tests an AI's understanding of number representation

DeepSeek R1
February 27, 2025
Text

?
Prompt

Is 9.11 greater than 9.9?

DeepSeek R1
Response

About DeepSeek R1

Capabilities

ConversationReasoningCode GenerationAnalysis

Categories

TextCodeMultimodal

Specifications

Provider
Deepseek
Released
2025-02-01
Size
LARGE
Parameters
671B total (37B active MoE)
Context
128,000 tokens