More

est · 2026-04-22T13:19:58 1776863998

China had many >1M KM electric taxies. Usually they degrade after 200,000 km, and they are like 2 generations behind the latest ones.

est · 2026-04-22T02:08:56 1776823736

Now someone please revive Microsoft Encarta ...

est · 2026-04-21T08:20:09 1776759609

https://news.ycombinator.com/from?site=openclaw.ai

hot damn

finger · 2026-04-21T22:18:56 1776809936

Wow. I’m curious as to why so much spam.

est · 2026-04-21T05:50:32 1776750632

installed since last HN post. So Bonsai (1-bit) and Ternary-Bonsai are different?

Can it be run on browsers with WASM/WebGPU?

xenova · 2026-04-21T14:32:43 1776781963

yep! :) https://huggingface.co/spaces/webml-community/bonsai-ternary...

est · 2026-04-20T09:15:15 1776676515

> I look at the list of contributors

Specifically if those avatars are cute animie girls.

tomaytotomato · 2026-04-20T09:37:35 1776677855

> Specifically if those avatars are cute anime girls.

I know you are half joking/not joking, but this is definitely a golden signal.

GaryBluto · 2026-04-20T10:02:04 1776679324

Positive or negative to you? Whenever I see more than one anime-adjacent profile picture I duck out.

pezgrande · 2026-04-20T10:58:42 1776682722

Positive ofc, most of them a top-tier Rust devs.

est · 2026-04-20T02:17:53 1776651473

like 1973 oil crisis? End of V8 engines (pun intended)

cbdevidal · 2026-04-20T05:36:20 1776663380

Yeah, like that. Modern engines eclipsed pre-1970s engines in performance, efficiency, and even (yes I said it) in reliability.

At a cost of simplicity and beauty. And two lost decades of mediocre performance. Sigh

est · 2026-04-17T16:33:02 1776443582

> how homogenous the internet has become in terms of design

I think it's because Steve Jobs killed Flash.

est · 2026-04-17T02:28:53 1776392933

> bootstrap an Artifacts repo from an existing git repository

wow that's cool. I used to hack CF Worker to operate .git using isomorphic-git, it's a PITA.

> ArtifactFS runs a blobless clone of a git repository: it fetches the file tree and refs, but not the file contents. It can do that during sandbox startup, which then allows your agent harness to get to work.

That's insanely useful. How combining with only committing the file changed we'd have a single-blob editing capability against any .git

est · 2026-04-16T16:34:39 1776357279

I really want to know what does M, K, XL XS mean in this context and how to choose.

I searched all unsloth doc and there seems no explaination at all.

tredre3 · 2026-04-16T20:14:56 1776370496

Q4_K is a type of quantization. It means that all weights will be at a minimum 4bits using the K method.

But if you're willing to give more bits to only certain important weights, you get to preserve a lot more quality for not that much more space.

The S/M/L/XL is what tells you how many tensors get to use more bits.

The difference between S and M is generally noticeable (on benchmarks). The difference between M and L/XL is less so, let alone in real use (ymmv).

Here's an example of the contents of a Q4_K_:

    S
    llama_model_loader: - type  f32:  392 tensors
    llama_model_loader: - type q4_K:  136 tensors
    llama_model_loader: - type q5_0:   43 tensors
    llama_model_loader: - type q5_1:   17 tensors
    llama_model_loader: - type q6_K:   15 tensors
    llama_model_loader: - type q8_0:   55 tensors
    M
    llama_model_loader: - type  f32:  392 tensors
    llama_model_loader: - type q4_K:  106 tensors
    llama_model_loader: - type q5_0:   32 tensors
    llama_model_loader: - type q5_K:   30 tensors
    llama_model_loader: - type q6_K:   15 tensors
    llama_model_loader: - type q8_0:   83 tensors
    L
    llama_model_loader: - type  f32:  392 tensors
    llama_model_loader: - type q4_K:  106 tensors
    llama_model_loader: - type q5_0:   32 tensors
    llama_model_loader: - type q5_K:   30 tensors
    llama_model_loader: - type q6_K:   14 tensors
    llama_model_loader: - type q8_0:   84 tensors

huydotnet · 2026-04-16T16:45:02 1776357902

They are different quantization types, you can read more here https://huggingface.co/docs/hub/gguf#quantization-types

arcanemachiner · 2026-04-16T22:26:57 1776378417

Just start with q4_k_m and figure out the rest later.

est · 2026-04-16T16:30:51 1776357051

hey you can do a bit research yourself and tell your results to us!