MotionVideo
0:15Comparison of DeepSeek models generating a pagoda
Comparison of DeepSeek models generating a pagoda — DeepSeek V4.1 Flash 1017 video by @WescheNex1q. Watch the clip and read the original creator’s post for context.
0:12 · 24 likes in the source snapshot · 1.3K views in the source snapshot
Creator post
Creator posts and original work titles are preserved in their source language.
Same prompt, same 2 DGX Sparks, same speed.
-DeepSeek V4.1 Flash: 46,355 tokens, 20.6 min
-V4.1 Flash 1017: 38,146 tokens, 16.9 min
18% fewer tokens for the same pagoda. Both 37.6 tok/s, thinking on, no output cap.
EXL3 2.9bpw quants by @Libertai_DAI (1017)
@deepseek_ai
This is the creator’s post text, not a verified generation prompt.
