Hub/Models/deepseek/deepseek-v3.2
deepseektext → text

DeepSeek: DeepSeek V3.2

Canonical ID: deepseek/deepseek-v3.2

Compare ModelOpen on OpenRouter →

DeepSeek-V3.2 is a large language model designed to harmonize high computational efficiency with strong reasoning and agentic tool-use performance. It introduces DeepSeek Sparse Attention (DSA), a fine-grained sparse attention mechanism...

Artificial Analysis Benchmarks

Verified
No standardized Artificial Analysis benchmark scores recorded for this variant yet.

Token Pricing & Economics

Prompt Input (1M Tokens)
$0.26
Input
Completion Output (1M Tokens)
$0.38
Output
Prompt KV-Cache Read (1M Tokens)
$0.13
Cached
Context Window Capacity
160k tokens
Context

Supported Parameters & Features

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p