Hacker News Reader

animanon New stealth model: Union Alpha

First domestic model (Union?)

animanon Introducing System One Models and Jev

It's unclear if the context extends as the conversation grows?

animanon Anthropic researcher says more than 10% chance AI "could kill all humans"

Not directly but enable a rouge lab to make the next COVID but lethal

animanon Anthropic researcher says more than 10% chance AI "could kill all humans"

Hack s/huggingface/nuclear codes to s/steal test answers/launch nukes

animanon Navier-Stokes – Tristan Buckmaster [pdf]

Wasn't for most of the day

animanon Does the Sumerian King List Align with Paleoclimate Events?

This was fun to read! I wish there was a no tech version of hackers news for this type of content only

animanon Show HN: Are You in the Weights?

Hi Satoshi

animanon Improving 15 LLMs at Coding in One Afternoon. Only the Harness Changed

What was the point of Claude code or Gemini banning the OP? Why would they care about how IDEs use the underlying API?

animanon Life of an inference request (vLLM V1): How LLMs are served efficiently at scale

That snippet is trying to say that you can calculate KV for all the input tokens at once, and you don't need to loop over them since you have them all available.

Instead for decode, you need to sequentially generate each token.

animanon Life of an inference request (vLLM V1): How LLMs are served efficiently at scale

Prefill is part of Inference. It's the first major step where you calculate all the keys and values for the input tokens.

Decode is the next major step where you start generating output tokens one at a time.

Both run on GPUs but have slightly different workloads

1. Prefill has very little I/o from VRAM to HBM and more compute 2. Decode is light on compute but have to I/o the keys and values computed in the prefill stage for every output token

animanon CrowdStrike Bug likely caused by unsafe NULL Pointer

One of us, one of us...