DeepSeek

DeepSeek V4 Flash 0731

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.

deepseek-v4-flash-0731

Pricing

~92%

Input: $0.003888 / 1,000,000 tokens
Output: $0.019669 / 1,000,000 tokens
Cache read: $0.002118 / 1,000,000 tokens
Cache write: $0.003888 / 1,000,000 tokens

How pricing works

Discount history

92.4%Nov 8, 03:56 AM UTC

Nov 7, 11:57 PMNov 8, 03:56 AM UTC

Model details

Input: text
Output: text
Context window: 1,048,576 tokens
Maximum output: 65,536 tokens
Availability: $2KUncapped capacity
Updated: 11/8/2026, 3:58:06 AM UTC

Capacity is reserved when your request is accepted.

Capabilities

ChatStreamingTool callingReasoningJSON mode

Supported parameters

frequency_penaltyinclude_reasoninglogit_biaslogprobsmax_tokensmin_ppresence_penaltyreasoningreasoning_effortrepetition_penaltyresponse_formatseedstopstructured_outputstemperaturetool_choicetoolstop_atop_ktop_logprobstop_p