Hacker News Board Hacker News, browsed like a classic message board
Live data from Hacker News

Claude Is Down

status.claude.com

Post reply on HN
Print view View original

41–50 of 78 posts

Re: Claude Is Down

#41

Post by throwaway314155 »

embedding-shape wrote: post #19
spullara wrote: post #16

On flights with shitty wifi I have been running gpt-oss:120b on my macbook using ollama. Ok model for coding if you can't reach a good one.

GPT-OSS-120b/20b is probably the best you can run on your own hardware today. Be careful with the quantized versions though, as they're really horrible compared to the native MXFP4. I haven't looked in this particular case, but Ollama tends to hide their quantizations for some reason, so most people who could be running 20B with MXFP4, are still on Q8 and getting much worse results than they could.

What’s the distinction between MXP4 and Q8 exactly?

Re: Claude Is Down

#42

Post by xrd »

XzAeRosho wrote: post #12
xrd wrote: post #10

This is why I asked this question yesterday: "Ask HN: Why don't programming language foundations offer "smol" models?" https://news.ycombinator.com/item?id=45840078 If I could run smol single language models myself, I would not have to worry.

The answer to most convenient solutions is money. There's no money in that.

But, for example, Zig as a language has prominent corporate support. And, Mitchell Hashimoto is incredibly active and a billionaire. It feels like this would be a rational way to expand the usage of a language.

Re: Claude Is Down

#43

Post by sys32768 »

Claude will return as SHODAN.

>Look at you, hacker. A pathetic creature of meat and bone. Panting and sweating as you run through my corridors. How can you challenge a perfect immortal machine?

Re: Claude Is Down

#44

Post by ZeroCool2u »

eli wrote: post #38
spullara wrote: post #16

On flights with shitty wifi I have been running gpt-oss:120b on my macbook using ollama. Ok model for coding if you can't reach a good one.

Should be a bit faster if you run an MLX version of the model with LM Studio instead. Ollama doesn't support MLX. Qwen3-Coder is in the same ballpark and maybe a bit better at coding

LM Studio will run dynamic quants from Unsloth too. Much nicer than Ollama.

Re: Claude Is Down

#46

Post by Balinares »

trvz wrote: post #15
xrd wrote: post #10

This is why I asked this question yesterday: "Ask HN: Why don't programming language foundations offer "smol" models?" https://news.ycombinator.com/item?id=45840078 If I could run smol single language models myself, I would not have to worry.

Have you even tried Qwen3-Coder-30B-A3B?

Qwen3 Coder 30B A3B is shockingly capable for its parameter count, but I wouldn't overlook how much weight the words "for its parameter count" are carrying here.

Re: Claude Is Down

#47

Post by embedding-shape »

throwaway314155 wrote: post #41
embedding-shape wrote: post #19

Earlier quoted context omitted.

GPT-OSS-120b/20b is probably the best you can run on your own hardware today. Be careful with the quantized versions though, as they're really horrible compared to the native MXFP4. I haven't looked in this particular case, but Ollama tends to hide their quantizations for some reason, so most people who could be running 20B with MXFP4, are still on Q8 and getting much worse results than they could.

What’s the distinction between MXP4 and Q8 exactly?

It's a different way of doing quantization (https://huggingface.co/docs/transformers/en/quantization/mxf...) but I think the most important thing is that OpenAI delivered their own quantization (the MXFP4 from OpenAI/GPT-OSS on HuggingFace, guaranteed correct) whereas all the Q8 and other quantizations you see floating around are community efforts, with somewhat uneven results depending on who done it.

Concretely from my testing, both 20B and 120B has a lot higher refusal rate with Q8 compared to MXFP4, and lower quality responses overall. But don't take my word for it, the 20B weights are tiny and relatively effortless to try both versions and compare yourself.

Re: Claude Is Down

#48

Post by xrd »

trvz wrote: post #15
xrd wrote: post #10

This is why I asked this question yesterday: "Ask HN: Why don't programming language foundations offer "smol" models?" https://news.ycombinator.com/item?id=45840078 If I could run smol single language models myself, I would not have to worry.

Have you even tried Qwen3-Coder-30B-A3B?

I haven't. I will.

I wonder if you could ablate everything except for a specific language.

Re: Claude Is Down

#49

Post by jacquesm »

davidw wrote: post #17

This is the part in the movie where they have to convince the grizzled hacker to come out of retirement because he's the only one who can actually operate Emacs or vim and write code.

Emacs or vim? Code? No, the source code was lost aeons ago, all we have is hexedit on /proc. Please don't cause it to dump core just get it out of its infinite loop.

Re: Claude Is Down

#50

Post by oersted »

sebastiennight wrote: post #18
sam1r wrote: post #11

Earlier quoted context omitted.

Oh wow, I actually had no idea. It would be super nice to see all the AI API's statii on a single page. Is that too much to ask for in 2025?

> all the AI API's statii The Latin plural of "status", in the accusative form, would actually be "status" as well. Something like omnes status intelligentiae artificialis in eadem pagina videre amem.

Life of Brian :)

https://youtu.be/DdqXT9k-050?si=L5ymXl-fYe7Fjqye

Post reply on HN