Skip to main content
DE

DeepSeek V3.2 Exp

deepseek/deepseek-v3.2-exp

DeepSeek-V3.2-Exp is an experimental large language model released by DeepSeek as an intermediate step between V3.1 and future architectures. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...

Compare
Context window
164K
Input
Text
Output
Text
Released
2025-09-29
Max output
66K
Knowledge cutoff
2025-07-31
Model parameters
685B
Weights
Open
License
mit
Capabilities
Open weight, Reasoning, Structured, Temperature, Tool call

Pricing

No official pricing available.

Estimate usage cost

Compare verified public prices for this exact model and region. Enter uncached input and cache usage separately. Estimates exclude unspecified fees, credits and currency conversion. Set requests to zero to estimate token charges only.

Performance

Overall capability

46.2/100

Coding
—
Reasoning
41.2
Professional work
—
Knowledge & language
56.7
CritPtDeepSeek V3.2 Exp (Non-reasoning)ReasoningArtificial Analysis2026-11-05
AA-LCRDeepSeek V3.2 Exp (Reasoning)Knowledge & languageArtificial Analysis2026-11-05
Humanity’s Last ExamDeepSeek V3.2 Exp (Reasoning)ReasoningArtificial Analysis2026-11-05
IFBench(Artificial Analysis)DeepSeek V3.2 Exp (Reasoning)Knowledge & languageArtificial Analysis2026-11-05
Arena Text OverallBlindLMArena2026-10-28
CritPtDeepSeek V3.2 Exp (Reasoning)ReasoningArtificial Analysis2026-10-27
AA-OmniscienceDeepSeek V3.2 Exp (Reasoning)Knowledge & languageArtificial Analysis2026-10-27
Arena Text OverallBlindLMArena2026-10-15

Providers

No providers found

ModelRegistry · Catalog updated 2026-11-05