Chinese startup DeepSeek unveils V4.1-Flash model
The initial unveiling.
5 items · 2 sources · 3 days
Operational story trace
Follow in this browser to see new updates on your Live feed.
Latest change
A Sep 12 AINews technical read calls the architecture leap substantial enough that "this should have been DeepSeek v5," a day after early reports of the release pressuring competitor API pricing.
DeepSeek unveiled V4.1-Flash on Sep 10, claiming it beats its own flagship V4-Pro while cutting the memory agents need to run it. By Sep 11 the release was already squeezing rivals on price.
State over time
The initial unveiling.
DeepSeek's own outperforms-V4-Pro claim.
First outlet to frame the release around agent memory costs specifically.
First sign the release is already moving competitor pricing.
AINews' technical deep dive frames the architecture as v5-caliber.
The initial unveiling.
First outlet to frame the release around agent memory costs specifically.
DeepSeek's own outperforms-V4-Pro claim.
First sign the release is already moving competitor pricing.
AINews' technical deep dive frames the architecture as v5-caliber.
What to watch — open questions
Storylines are threaded mechanically from the feed: stories that share a distinctive anchor across multiple days and sources. Each item links to its original source. The evidence trace, current state, and open questions are written by the editor routine and refreshed whenever a new beat lands.