Loading

Preparing your page

AI Stats is fetching the latest data for this page. This usually only takes a moment.

Return to the main page and start from a clean slate.

Explore the catalogue of AI models and their details.

Check the documentation for guides, references, and examples.

Explore the Gateway

Access the largest unified AI Gateway and use over 300+ models easily.

If this screen doesn't disappear after a short while, you can refresh the page or use one of the links above to continue.

DeepSeek V3 (2024-12-26) Benchmarks - Performance Metrics & Comparisons | AI Stats

DeepSeek V3 (2024-12-26)

DeepSeek V3 (2024-12-26)

DeepSeek

Overview Family Timeline Benchmarks Availability Pricing Gateway Performance

Highlights

Top benchmark results for deepseek/deepseek-v3-2024-12-25.

1016#12

0.39#34

5.16#13

0.59#46

1355#23

NYT Connections

0.15#27

0.19#30

Thematic Generalisation

2.03#16

Benchmark table

Detailed scores across tracked benchmarks.

Benchmark	Category	Top Score	Info	Self Reported	Source
Ai2 SciArena	-	1016	-	No	Source
AIME 2024	math	0.39	-	No	Source
Elimation Game	-	5.16	-	No	Source
GPQA Diamond	general-knowledge	0.59	-	No	Source
LMArena Text	-	1355	-	No	Source
NYT Connections	-	0.15	-	No	Source
SimpleBench	-	0.19	-	No	Source
Thematic Generalisation	-	2.03	-	No	Source

Benchmark comparisons

Use the selector to switch benchmarks and see how this model stacks up against its closest competitors.

Ai2 SciArena

Compare this model with the leading peers for the selected benchmark.

Benchmark

1016

Rank #12/16

16 models

Showing 11 models around the selected model (out of 16 total).

View benchmark page