DeepSeek V4.1 Flash is a sparse mixture-of-experts model from DeepSeek, and the first built on the company's Causal Encoder-Decoder (CED) architecture. It activates 8B parameters on input and 16B on...
Rank
#1
Up 4% week over week
Tokens
37.1T
2026-10-15
Requests
774.6M
774,616,833
Tool calls
-
Input / M
$0.04
Output / M
$0.66
Cache / M
$0.0054
Context
1.0M
1,048,576 tokens