Deepseek-v4.1-flash I'm really curious to understand the architecture behind the scenes.
🤔 Model is really fast as you can see in the video! I've got an average decoding speed of around 350 t/s. BTW it is clearly experimental and: - currently, the model is under heavy load and
Quite beautiful: DeepSeek V4.1 Flash hands-on 3D test, this Fuji Sakura station is already playable!
The low-model station building, pink cherry blossoms, railway tracks, and fences were all set up at once, and day and night switching, sunny and rainy, and animation pauses were all connected.
DeepSeek 4.1 Flash is SUPER FAST!
(Video) Even faster than my previous test. 300tok/s sustained???
Surprise landing of DeepSeek v4.1-flash.
I am blown away by the quality improvement compared to v4-flash. Huge jump. It is FAST peaked speed 427 t/s. It used 112k tokens and 20 turns taking 7:41. Cost 0.037 Unbelievable result for 4 cents. deepseek_ai is cooking.
You have reached the end of the archive
All of deepseek