DeepSeek V4 Flash 0731

DeepSeek · deepseek/deepseek-v4-flash-0731

← Back to leaderboard

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.

open weightstexttext->text

Context

Max context: 1048576
Max output: 384000

Pricing

Input / 1M: 0.14
Output / 1M: 0.28
Blend / 1M: 0.21

Quality

Quality index:

Provider

Provider: DeepSeek
Moderated: no