Dusted

Midas list top quartile contrarian superforecaster

[Dusted @shulatrenches]

Shorting some $ZEC

[Dusted @shulatrenches]

Let’s try a $LIT long

[Dusted @shulatrenches]

This is the zeitgeist

[Dusted @shulatrenches]

@PaulSkallas I lived in the building this vid is pointing at and was on my balcony every night

[Dusted @shulatrenches]

@5morewhooppers All good man it’s just a string of letters and numbers 👍

[Dusted @shulatrenches]

@cartaleoni 💎

[Dusted @shulatrenches]

@BarryOnHere 😂

[Dusted @shulatrenches]

@bryan_johnson Let’s go Bryan

[Dusted @shulatrenches]

@capitalatrisk23 Cope

[Dusted @shulatrenches]

@based16z 15M a year to shill polymarket. Brons agent is generationally brainded

[Dusted @shulatrenches]

I actually can’t wrap my head around the implications of 67X inference performance on nvidia Rubin

[Dusted @shulatrenches]

Difficult game sometimes. We try again tomorrow

[Dusted @shulatrenches]

@aptosgooner Objective and balanced take

[Dusted @shulatrenches]

Nvm !

[Dusted @shulatrenches]

Long ZEC I guess

Am I doing this right ?

[Dusted @shulatrenches]

@reaIityobserver chills

[Dusted @shulatrenches]

@swansonbenson0 lmk

[Dusted @shulatrenches]

Long ZEC I guess

Am I doing this right ?

[Dusted @shulatrenches]

We can speculate all day on how much we use AI in our meager daily lives, but on the supply side, demand is THERE. Many likely psyop themselves into thinking AI isn’t useful to THEM, but the tokens continue to go brrrr

[SemiAnalysis @SemiAnalysis_]

AGENTIC TRAFFIC NOW MAKES UP MORE THAN 70% OF ALL INFERENCE TRAFFIC 🚀

Agentic workloads are characterized by four elements:

🟠 Multi-turn: a session includes tens or hundreds of turns, leading to high potential KV-cache reuse.

🟠 Long context: system prompts, tool definitions, and the large number of turns make context accumulate quickly.

🟠 High prefix reuse: since the conversation progresses linearly, where output from turn n-1 is concatenated to turn n (typically), most context can be served from KV cache rather than recomputed (this depends on the amount of storage available to store KV tensors). As n grows, the ratio of cached input relative to uncached input typically tends towards 1.

🟠 Sub-agent bursts: a session launches multiple short-lived sub-agents with fresh context, which create bursty KV-cache patterns.

[Dusted @shulatrenches]

@rennyzucker @leevalueroach Cool