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
#202
Up 129% week over week
Tokens
8.1B
2026-10-21
Requests
1.4M
1,366,703
Tool calls
-
Input / M
$0.05
Output / M
$0.15
Cache / M
$0.004
Context
1.0M
1,048,576 tokens