Back to all models
deepseek
deepseekDeepSeek: DeepSeek V4 Pro
DeepSeek V4 Pro is a large-scale Mixture-of-Experts model from DeepSeek with 1.6T total parameters and 49B activated parameters, supporting a 1M-token context window. It is designed for advanced reasoning, coding,...
deepseek/deepseek-v4-pro
Context Size
1,048.576K
Input Price
43,500 Ks/M
Output Price
87,000 Ks/M
Architecture
Text
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Details
TokenizerDeepSeek
Max Completion384,000 tokens
Provider Context1,048.576K tokens
ModeratedNo
CreatedApr 24, 2026