Hacker Newsnew | past | comments | ask | show | jobs | submit | jasongill's commentslogin

In case anyone missed it earlier on HN, there is a similar project that runs on an ESP32: https://github.com/hchunhui/tiny386

That one is offering Pentium Pro and SSE3 emulation. Not too shabby.

"Add customer delight to the product pages; make no mistakes"

Try it in a private browsing window - if you are logged in, Google will link directly to the result, but if you aren't logged in, it appears to be redirecting through their /goto endpoint

"© 2026 lazygit.dev — Unofficial site. Not affiliated with the official lazygit project or Jesse Duffield."

strange


Same as with the tmux.app website. I looked around and I don't understand who's making these vibecoded websites.


Mostly, the ones I see are from SEO scammers and domain parkers trying to pump the value of their domains to resell them, or straight up scamming others to download or even buy a re-packaged malware version of their open source software.

This is likely the reason why they do this and now vibecoding makes it easy for anyone to do this.


I believe doomemacs.com has nothing to do with the developers either. Just more SEO spam or something.


This is something I’ve seen before. Someone is registering domain names related to open source projects and then ‘creating’ slop websites about the projects. Not sure what the play is, or whether this falls into that category, but it’s really weird and takes control away from open source maintainers in a scary way. Maybe it’s extortion or maybe it’s an malware thing. I highly doubt it’s out of the goodness of someone’s heart.

Open source maintainers shouldn’t have to register their projects as trademarks/proactively register domains to avoid this, but that seems like the only strategy?


Best case scenario they’ll try to monetize the site. Worse case it’s a malware distribution play (can eventually change legit repo link to malicious repo)


It would be great if they made their inference capacity for this model available via OpenRouter; the fastest provider on OpenRouter right now is at ~80tps https://openrouter.ai/qwen/qwen3.8-27b#providers

They do appear to host other models on OpenRouter so maybe Qwen3.8 will be there soon: https://openrouter.ai/provider/cerebras


We're serving it around 150-200tok/s (uses our new speculative decoding implementation on a DFlash2 draft model).

https://mixlayer.com, LAUNCH-Q38-27B gets you $5 in credits if you want to kick the tires.


I tried in your playground and got 14.2 tok/s?


apologies we just got a sudden burst of new users and traffic, it's scaling up now.


just added 8 more H200s to the cluster, if you (or anyone else) runs into issues please feel free to drop me a message: zack at mixlayer.com


Works much better now! Got 103.9 tok/s, not quite 200 - but still amazing! Thanks for sharing


Something a lot of model providers don't talk about: any time an engine uses speculative decoding the throughput will depend on how much your output token distribution matches what the draft model was trained on.

The DFlash2 draft model we're using was trained on a lot of code, so if you use it in a coding agent you'll probably notice it run a lot faster (we've seen it break 300 tok/s).


FYI, I might be missing something but I think your billing system might not be working well - I'm not seeing any indication in the UI that my usage is being deducted from the $5 of free credits.


Hey Daniel! It's a bit hidden, but at the bottom of the billing page there's a "Credits" section which should show usage of any active credits and the balance remaining. The usage/billing metrics are batched/handled async so it might take a minute or so for usage to be reflected. Let us know if it feels off.


I don't see any kind of input cache discount listed on your pricing page. Do you offer that, or is all input priced the same?


Trying now. History question: got 96 t/s. Code review task: 130 t/s. Nice!


any way to see the tok/s for all the models listed on your homepage? curious which has the best speed/quality tradeoff for me


This feels great


It appears that they do support Prompt Caching: https://inference-docs.cerebras.ai/capabilities/prompt-cachi...


It doesn't reduce the price though.


Good catch, I guess I got lost in the marketing speak of the page!


> How are cached tokens priced?

> There is no additional fee for using prompt caching. Input tokens, whether served from the cache or processed fresh, are billed at the standard input token rate for the respective model.

Well, talk about flipping the narrative.


heh

Is there a speed increase or is that purely marketing spin on “we might cache on our end but no discount for you”?


Pure marketing.


It's agents all the way down


Feel free to submit a pull request


Yes, I think this project is a good example of how complex domains aren't simply solvable just by pointing ever more intelligence at the system, even when the code is all open source and the issues are all well defined. I say this as a contributor!


I did and no one did review despite me begging on the vllm slack!


And honestly? That changes the game.


As soon as I saw "Quake Shareware" in the title of this post, I thought "Pray to the one you will pay!" which was what appeared on the screen when you ran QCRACK.

I always wondered why Final Doom didn't unlock with QCRACK - I always assumed it was a bug with the keygen, never realized it wasn't possible to purchase it from the disk at all!


You can actually unlock Final DOOM from qcrack. But you need to use `-g` to give the codename of the game[1] because sku.17 won't give it away.

1. https://rmolina-co.blogspot.com/2016/08/qcrack-y-la-belleza-...


Guidelines | FAQ | Lists | API | Security | Legal | Apply to YC | Contact

Search: