WHAT NEW IN X10.29 THU

AI collections from X · 22 collections · 60,637 posts

Back to No.020 DeepSeek V4.1 Flash hands-on tests and comparisonsUpdated 10 hours ago
deepseek-v4.1POST

DeepSeek V4.1 Flash reads a 1M-token prompt in 29s on DGX Station

A single 1M-token request returned the correct answer in 29 seconds, with shorter prompts read at about 43K tok/s.

Garner Fox McCloud
@GarnerFoxArchived 11 days agoOriginally posted 2026.10.17
Garner Fox McCloud @GarnerFoxOriginally posted 2026.10.17
Open original on X

How long does my DGX Station take to read a million tokens? About 29 seconds. DeepSeek V4.1 Flash 1017, one request, a 1M token prompt, right answer back in 29s. Shorter prompts read at about 43K tok/s.

x.com/garnerfox/status/2111477343328100440Archived 2026.10.18 16:43
Open original on X