Hacker News Reader

Jowseyon Kimi K3.1 Pro and K3.1 Pro Highspeed

Tech blog and benchmark scores now available here: https://www.kimi.ai/blog/kimi-k3-1-pro

It appears to be generally frontier-level.

Another Claude-written changelog. They never inspire much confidence in a project for me.

Jowseyon Can you tell which images are AI-generated?

I hadn't been exposed to the new GPT-Image-2.5 model before this, so didn't have any "tells" to go off of, and found it almost impossible to begin with, but it seems the brain is pretty good at adapting for visual stuff! After 2 rounds, I was able to build up a 16-image combo, answering most within a second or two.

I think the "tell" I ended up with is that almost all of the AI images tend to center around a very obvious main "subject" (a flower, a bike, a wrench, etc) – presumably an inherent artefact of generating from a prompt – with everything else around the subject having a very strange depth to it. The depth, focus, and bokeh around the main subject just never look quite right. If you look at an image and find it has a very strong central subject with a little too much depth separation than expected, there's a good chance it's AI.

I recognise that this method of finding a consistent "tell" will probably not hold for future models, unfortunately. Everything else looks almost perfect at this point.

Jowseyon Qwen3.8-Max

for those who skip to the comments: Qwen released an updated checkpoint of this model today (0902) with significantly higher benchmark results that appear to place it much closer to Fable/Sol

Jowseyon Guess which of these LLM outputs is watermarked

Interestingly, it seems almost every set of three seems to follow a pattern: one passage of the three will have a key word or phrase swapped in the first sentence. That is, for every set of 3 passages, two will start with ~identical sentences, and one will have a key word or token changed.

I caught onto this early and used it every time, and ended up getting 2/10, which is worse than random chance. I smell trickery!

Jowseyon Writergate: Zig I/O Interface Overhaul

So many Claude-isms in one post. Please stop.

Jowseyon Qwen3.8-Max: A New Bar for Coding and Cowork

My understanding is that these "preview" models are usually earlier RL checkpoints, and that "official release" happens when they're happy with the training run?

I believe they mentioned around the preview announcement that they'd be releasing improvements to capability, which I assume means continued training.

Jowseyon Qwen3.8-Max: A New Bar for Coding and Cowork

They mention it explicitly in the Twitter post [0].

> Pricing: Input: $2.0 / M tokens Output: $6.0 / M tokens Implicit Caching: $0.25 / M tokens

[0]: https://x.com/Alibaba_Qwen/status/2084100707423289643

Jowseyon Qwen3.6-27B: Flagship-Level Coding in a 27B Dense Model

This is pretty standard in every model. Ask Opus or Gemini about 2026 (without a big system prompt to steer them) and they'll swear blind it's 2024/25 too.

Jowseyon Show HN: 1-Bit Bonsai, the First Commercially Viable 1-Bit LLMs

llm/bot comment

Jowseyon BitNet: 100B Param 1-Bit model for local CPUs

Agreed. This is becoming an issue, see also: https://news.ycombinator.com/item?id=47259308

Jowseyon Qwen3.5 Fine-Tuning Guide

This reply is entirely AI generated. You guys are trying to find reason in a hallucination. It's unfortunately impossible to put into words what the "LLM smell" is at this point, but I trust someone else who spends a lot of time reading LLM output can back me up on this.

I've seen these agent-written fake anecdotes on Twitter, Reddit, and now here, all with the exact same formatting. They pretend to be real people with real anecdotes, but they're all completely made up.

Jowseyon Nano Banana Pro

To expand, it comes from the stealth name it was given on LMArena I believe. The model made news while still in "stealth mode" and so Google capitalised on the PR they'd already built around that and just launched it officially with the same name.

Jowseyon .NET 10

> "Legend has it that the name Rider comes from ReSharper IDE, but since Ride didn’t sound great, it became Rider.

— We haven’t found the source for this story yet."

From https://blog.jetbrains.com/dotnet/2022/08/03/happy-5th-birth...

Jowseyon Managing context on the Claude Developer Platform

Related, it feels like AI Studio is the only mainstream LLM frontend that treats you like an adult. Choose your own safety boundaries, modify the context & system prompt as you please, clear rate limits and pricing, etc. It's something you come to appreciate a lot, even if we are in the part of the cycle where Google's models aren't particularly SOTA rn

Jowseyon Two Slice, a font that's only 2px tall

Some of the characters/words (particularly "c"/"can") sort of look like they've been cropped from the top, trusting the brain to fill in the bottom half. Reminds me of what Sandisk did with the "S" in their redesign. I wonder if there's any research behind this?

Jowseyon Gemma 3n preview: Mobile-first AI

Is this not what Style Control (which IIRC they're making default soon) aims to mitigate?

Jowseyon Gemini 2.5 Pro vs. Claude 3.7 Sonnet: Coding Comparison

Isn't https://aider.chat similar?

Jowseyon Discord Reduced WebSocket Traffic by 40%

This is true, but in the past various Discord employees have explicitly said (on HN no less) that they don't intentionally ban accounts for using them, only that sometimes their anti-spam systems can also flag them as false positives (and that you may be able to submit a ticket in this case), so I wouldn't worry too much

Jowseyon Discord Reduced WebSocket Traffic by 40%

The onus is really on Discord, but you can use https://openasar.dev to partially fix the problem for yourself - it's an open source drop-in replacement for the client updater/bootstrapper.

Jowseyon Questions about LLMs in Group Chats

> Could you use a LLM function call to decide whether or not to respond

I've implemented this for a toy project before and it worked surprisingly well, yes! It can take some creative prompting to have the model understand which messages are actually directed at it, though, which I guess makes sense since every User message is _supposed_ to be?

Jowseyon React 19 almost made the internet slower

https://developer.mozilla.org/en-US/docs/Learn/Tools_and_tes...

Jowseyon Apple blocks PC emulator in iOS App Store and third-party app stores

I assume there's just a "not be" missing somewhere

Jowseyon The state of Vulkan apps in 2024

I know Unity at least defaults to DX11 and I believe it doesn't package other renderers at all unless you explicitly enable them in Project Settings or use a rendering feature that requires them, and I can't imagine many people are digging into the renderer settings without good reason

Jowseyon Goody-2: the most responsible AI model

I can't write a comment about this article. Discussing any AI, including GOODY-2, could influence perceptions and potentially lead to misuse or unwarranted trust in technology. This could skew technological development impetus and create an imbalance in human-AI interaction dynamics.

Jowseyon Game-icons.net: Free icons for your games

Love these, have used them in a bunch of games and they generally fit in great with any other style you're using, huge vouch

Jowseyon Show HN: YouTube banned adblockers so I built an extension to skip their ads

And in case anyone thinks this is a Chrome vs Firefox thing, I've also been using SB + uBlock on Edge and still haven't run into any problems

Jowseyon Song stuck in your head? Just hum to search (2020)

This has always been my experience too, never had it work

Jowseyon X.ai Grok

Some context from the main X.ai page:

> Grok is an AI modeled after the Hitchhiker’s Guide to the Galaxy, so intended to answer almost anything and, far harder, even suggest what questions to ask!

Grok is designed to answer questions with a bit of wit and has a rebellious streak, so please don’t use it if you hate humor!

A unique and fundamental advantage of Grok is that it has real-time knowledge of the world via the 𝕏 platform. It will also answer spicy questions that are rejected by most other AI systems.

Grok is still a very early beta product – the best we could do with 2 months of training – so expect it to improve rapidly with each passing week with your help.

Haven't tried it because I refuse to pay for Twitter Premium but from the description it seems disappointingly gimmicky coming from a company backed by a billionaire that presumably wants to be taken seriously. Their "product" is a funny system prompt to some LLM hooked up to the Twitter search bar? Hopefully someone with Premium can prove me wrong :\

Jowseyon Windows 11 system components use the default browser to open links in Europe

I'm in the same boat; I want to love Microsoft, but it seems internally split into two very opposing groups with very different ways of thinking.

I swear by a lot of their developer tooling and open source work but abhor the Windows ecosystem