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....
In / out price$0.016/$0.64per 1M tokens
Context
1M tokens
Parameters
284BMoE13B active · 6/256 experts per token
Modalities
→
Released
Jul 31, 2026
2 up · 0 downCompare with another model
Net score over time
Oct 14 · +2Reviews (0)
Sign in to vote and write a review.
No reviews yet — be the first to review DeepSeek V4 Flash 0731.