Skip to content
World clockEU--:--UK--:--USA--:--CN--:--PLDEFRIT中文EN

portal about AI and technologyevents · analysis · interviews · technical background

Search
LIVE
›

FLASH24 English

page 163 / 171

Latest

19 texts
01

KV cache at 890 bytes per token: how DeepSeek squeezed attention memory

The model card lists 890 bytes of cache per token, about a quarter of what V4-Flash uses, and HBM demand at a quarter of the previous generation.

AI & models26 September 20265 min read
02

DeepSeek V4.1 Flash: 552 billion parameters and an asymmetric encoder-decoder

Released on 10 September 2026: a multimodal Mixture-of-Experts with 552 billion parameters activates 8 billion per input token and 16 billion per output token. The architecture changed, not just the scale.

AI & models26 September 20266 min read
03

AI Act: Europe enforces the rules while the UN seeks a global standard

Europe's AI regulation is the world's first complete legal framework, built on four risk levels and nine banned practices. In New York, the talk turned to an international coalition for standards.

Technology26 September 20265 min read
05
08
09
11

Tesla's Optimus V3 production stumbles on hands and AI

Technology26 September 20266 min read
14
16
18

Alibaba unveils Zhenwu V900 chip and opens T-Head SAIL stack

Technology25 September 20266 min read