What is DeepSeek V3.2-Speciale?

DeepSeek V3.2-Speciale is a large language model from DeepSeek, released on December 1, 2025. It accepts text input.

When was DeepSeek V3.2-Speciale released?

DeepSeek released DeepSeek V3.2-Speciale on December 1, 2025.

How does DeepSeek V3.2-Speciale score on benchmarks?

DeepSeek V3.2-Speciale has 9 cited benchmark scores, including WeirdML (Håvard Tveit Ihle) 46.7%, SimpleBench (SimpleBench) 52.6% and AIME 2025 96%. Each score links to its source.

DeepSeek V3.2-Speciale

DeepSeek·Operational·Released 2025-12-01text

Context-

Max output-

Input / 1M-

Output / 1M-

Cached in-

Benchmarks9

Benchmarks

Cited public sources · rank is exact-benchmark, apples-to-apples

WeirdML (Håvard Tveit Ihle)

#51/9246.7% ↗

SimpleBench (SimpleBench)

#47/7352.6% ↗

AIME 2025

#4/2496% ↗

GPQA Diamond

#39/8385.7% ↗

HLE (text-only)

#1/330.6% ↗

HMMT Feb 2025

99.2% ↗

HMMT Nov 2025

94.4% ↗

IMOAnswerBench

#3/384.5% ↗

LiveCodeBench (Pass@1-COT)

88.7% ↗

Compare

vs Gemini 4 Argon vs GPT-6.1 Sol vs Claude Sonnet 5.5 vs Muse Spark 1.3 vs GPT-6 Sol vs GPT-6 Luna vs Claude Opus 5.5 vs GPT-6 Astra

More from DeepSeek

DeepSeek V4.1 Flash DeepSeek V4 Pro DeepSeek V3.2 DeepSeek R1 DeepSeek V3 DeepSeek V4 Flash

DeepSeek V3.2-Speciale FAQ

What is DeepSeek V3.2-Speciale?: DeepSeek V3.2-Speciale is a large language model from DeepSeek, released on December 1, 2025. It accepts text input.
When was DeepSeek V3.2-Speciale released?: DeepSeek released DeepSeek V3.2-Speciale on December 1, 2025.
How does DeepSeek V3.2-Speciale score on benchmarks?: DeepSeek V3.2-Speciale has 9 cited benchmark scores, including WeirdML (Håvard Tveit Ihle) 46.7%, SimpleBench (SimpleBench) 52.6% and AIME 2025 96%. Each score links to its source.

Access DeepSeek V3.2-Speciale and every other model through one endpoint with automatic failover: Respan gateway.