DeepSeek V4 Flash 0731 (batch)
by DeepSeektext->text2 endpoints
Overview
DeepSeek V4 Flash 0731 is a sparse mixture-of-experts model from DeepSeek, with 13B active parameters out of 284B total. This re-post-trained revision is suited for coding, reasoning, and agent workflows. This is the GA release of DeepSeek V4 Flash.
Capabilities
Text generation
Tool use / Function calling
Structured output
Extended reasoning
Prompt caching
Long context
Top-P sampling
Stop sequences
Log probabilities
Modalities
Input
Text
Output
Text
Technical Specifications
Context Window
1.0M tokens
Max Output
943.7K tokens
Tokenizer
DeepSeek
Supported Parameters
frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatstopstructured_outputstemperaturetool_choicetoolstop_ktop_logprobstop_p
Pricing
Per 1M tokens
Input$0.11
Output$0.33
Cache read$0.0035
Added
July 31, 2026
Last synced 9/20/2026