ModelWatch
← Catalog

DeepSeek V4.1 Flash

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...

visionlong-contextopen-weightslow-cost
Input price
$0.05 / 1M tokens
Output price
$0.20 / 1M tokens
Context window
1.05M tokens
Max output
393K tokens
Output speed
216 tokens/s
Time to first token
0.99s
Inputs
text, image
Open weights
Yes
OpenRouter id
deepseek/deepseek-v4.1-flash
Listed
Sep 10, 2026

Benchmarks

Intelligence Index
n/a
Engineering Index
n/a

Speed and benchmarks for “DeepSeek V4.1 Flash (Max)” from Artificial Analysis. Prices, context window and listing date are live from OpenRouter.

Compare with others