Back

DeepSeek

DeepSeek V3.2 Speciale

ReasoningOpen weightsHugging Face
Intelligence
14.5
Rank 317 of 744 measured models
Context window
128K
Rank 773 of 1199 measured models

DeepSeek V3.2 Speciale scores 14.5 on the Artificial Analysis Intelligence Index, ranking 317th of 744 measured models. The closest models on this index are Grok 3 mini Reasoning (High) (14.6), Gemini 2.5 Pro Preview (May' 25) (14.5), and K-EXAONE (14.4).

Artificial Analysis Indices
Intelligence14.5
Rank 317 of 744 measured models
Math96.7
Rank 4 of 270 measured models
Standard Benchmarks
MMLU Pro86.3%
Rank 17 of 349 measured models
GPQA Diamond87.1%
Rank 108 of 619 measured models
HLE28.7%
Rank 161 of 664 measured models
LiveCodeBench89.6%
Rank 3 of 343 measured models
AIME 202596.7%
Rank 4 of 270 measured models
IFBench63.9%
Rank 114 of 450 measured models
AA-LCR70.0%
Rank 218 of 575 measured models
TerminalBench Hard34.8%
Rank 83 of 432 measured models
Models at a similar level
The closest models on the Intelligence Index, one configuration per family.

Preferences and analytics

Rybbit: visits, performance, clicks, copies, forms and session replay (masked inputs), with your permission. Refusing does not limit the site.

Privacy

Allow navigation analytics?

With your permission, Rybbit measures visits, performance and interactions (clicks, copies, forms), including session replay, to improve BenchSift. Input values are masked in replays. Refusing does not limit the site.

Your choice is stored for 180 days. Change it at any time using “Privacy preferences” in the footer. Withdrawal reloads the page.

No choice stored: Rybbit is disabled.

Privacy