DeepSeek V3 (03/24/2025)

Release Date: 3/24/2025

Benchmarked by Vals Logo

Avg. Accuracy

67.6%

Latency

36.2s

Cost (In/Out)

1.20 / 1.20

Context Window

131k

Max Output Tokens

131k

Input Modality

Benchmarks

Accuracy

Rankings

CorpFin

0.0%

58/ 58

CaseLaw

0.0%

40/ 40

TaxEval

0.0%

74/ 74

Math500

0.0%

54/ 54

AIME

0.0%

64/ 64

MGSM

0.0%

66/ 66

LegalBench

0.0%

89/ 89

MedQA

0.0%

70/ 70

GPQA

0.0%

66/ 66

MMLU Pro

0.0%

64/ 64

LiveCodeBench

0.0%

63/ 63

IOI

0.0%

21/ 21

Academic Benchmarks
Proprietary Benchmarks (contact us to get access)

Join our mailing list to receive benchmark updates

Model benchmarks are seriously lacking. With Vals AI, we report how language models perform on the industry-specific tasks where they will be used.

By subscribing, I agree to Vals' Privacy Policy.