Models

DeepSeek R1 0528 (May '25)

DeepSeek R1 is a reasoning-focused model from the R1 series, optimized for complex tasks requiring step-by-step thinking. It excels in mathematics, coding, and logical reasoning by leveraging advanced reinforcement learning and chain-of-thought methodologies.

ReasoningCodingCheap
Input / 1M tokens
$1.35
Output / 1M tokens
$4.20
Supported plans
14

Benchmark history

Evaluations

15

TAU2

Measured May 14, 2026Source

Score

0.37

Terminalbench Hard

Measured May 14, 2026Source

Score

0.16

Lcr

Measured May 14, 2026Source

Score

0.55

Ifbench

Measured May 14, 2026Source

Score

0.4

Aime 25

Measured May 14, 2026Source

Score

0.76

Aime

Measured May 14, 2026Source

Score

0.89

Math 500

Measured May 14, 2026Source

Score

0.98

Scicode

Measured May 14, 2026Source

Score

0.4

Livecodebench

Measured May 14, 2026Source

Score

0.77

Hle

Measured May 14, 2026Source

Score

0.15

Gpqa

Measured May 14, 2026Source

Score

0.81

Mmlu Pro

Measured May 14, 2026Source

Score

0.85

Artificial Analysis Math Index

Measured May 14, 2026Source

Score

76

Artificial Analysis Coding Index

Measured May 14, 2026Source

Score

24

Artificial Analysis Intelligence Index

Measured May 14, 2026Source

Score

27.1

Plan availability

Products and plans that support this model

5

Discussion

Thinking... Make sure you are connected to GitHub server