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

Hetzner is working on LLM Inference

sliplane.io

Post reply on HN
Print view View original

71–80 of 92 posts

Re: Hetzner is working on LLM Inference

#71

Post by nubg »

petesergeant wrote: post #49
nubg wrote: post #24

Earlier quoted context omitted.

nobody writes like that. the article is based on a prompt, which i would much rather read, instead of the ironed out version that the llm produced

> the article is based on a prompt, which i would much rather read I would much rather read someone's carefully written and LLM-assissted article than shallow and lazy dismissals like this one. What, precisely, do you feel you've added to the conversation here?

Why not publish the prompt anyway and let readers decide if they want to run it through an LLM (maybe next year's LLM, which will probably be better)?

Re: Hetzner is working on LLM Inference

#72

Post by atherton94027 »

swiftcoder wrote: post #62
atherton94027 wrote: post #39

Earlier quoted context omitted.

Never understood why hetzner gets brought up so positively here when there are dozens of EU competitors who are just as good and don't run 7 year old discounted servers

> don't run 7 year old discounted servers That's a pretty big part of why, honestly. How many of us are running server workloads that actually need the price/performance tradeoff of cutting edge server hardware?

We're not even talking cutting edge. If you browse hetzner's auction page you can see they're renting machines with almost decade old CPUs like the Xeon W-2145 or i7 6700. That's just a lot of old hardware..

Re: Hetzner is working on LLM Inference

#73

Post by petesergeant »

nubg wrote: post #71
petesergeant wrote: post #49

Earlier quoted context omitted.

> the article is based on a prompt, which i would much rather read I would much rather read someone's carefully written and LLM-assissted article than shallow and lazy dismissals like this one. What, precisely, do you feel you've added to the conversation here?

Why not publish the prompt anyway and let readers decide if they want to run it through an LLM (maybe next year's LLM, which will probably be better)?

What do you think this magical prompt will be, exactly, that perfectly represents what was almost certainly an iterative rather than spot process on the author's part?

Re: Hetzner is working on LLM Inference

#74

Post by beernet »

swiftcoder wrote: post #7

It would certainly be interesting to have a highly respected EU-native inference provider, if only to make the regulatory gods happy

To all mentions of IONOS / StackIT / OVH:

All of these are lightyears behind US/China LLM offerings. None of them offer any model close to open-source SOTA. Runtime and reliability is a disaster and sure enough you have to pay a "sovereignty" mark up.

Re: Hetzner is working on LLM Inference

#75

Post by jonas_scholz »

petesergeant wrote: post #73
nubg wrote: post #71

Earlier quoted context omitted.

Why not publish the prompt anyway and let readers decide if they want to run it through an LLM (maybe next year's LLM, which will probably be better)?

What do you think this magical prompt will be, exactly, that perfectly represents what was almost certainly an iterative rather than spot process on the author's part?

also slightly disrespectful to think that this was a single prompt, but ok:D

Re: Hetzner is working on LLM Inference

#76

Post by petesergeant »

jonas_scholz wrote: post #75
petesergeant wrote: post #73

Earlier quoted context omitted.

What do you think this magical prompt will be, exactly, that perfectly represents what was almost certainly an iterative rather than spot process on the author's part?

also slightly disrespectful to think that this was a single prompt, but ok:D

Exactly that

Re: Hetzner is working on LLM Inference

#77

Post by idiotsecant »

atherton94027 wrote: post #72
swiftcoder wrote: post #62

Earlier quoted context omitted.

> don't run 7 year old discounted servers That's a pretty big part of why, honestly. How many of us are running server workloads that actually need the price/performance tradeoff of cutting edge server hardware?

We're not even talking cutting edge. If you browse hetzner's auction page you can see they're renting machines with almost decade old CPUs like the Xeon W-2145 or i7 6700. That's just a lot of old hardware..

So? If it's cheaper and gets the job done who cares?

Re: Hetzner is working on LLM Inference

#78

Post by expedition32 »

beernet wrote: post #74
swiftcoder wrote: post #7

It would certainly be interesting to have a highly respected EU-native inference provider, if only to make the regulatory gods happy

To all mentions of IONOS / StackIT / OVH: All of these are lightyears behind US/China LLM offerings. None of them offer any model close to open-source SOTA. Runtime and reliability is a disaster and sure enough you have to pay a "sovereignty" mark up.

Freedom isn't free. Unless you want to live in a world were your children have to read the bible in class sacrifice has to be made.

Re: Hetzner is working on LLM Inference

#79

Post by beernet »

expedition32 wrote: post #78
beernet wrote: post #74

Earlier quoted context omitted.

To all mentions of IONOS / StackIT / OVH: All of these are lightyears behind US/China LLM offerings. None of them offer any model close to open-source SOTA. Runtime and reliability is a disaster and sure enough you have to pay a "sovereignty" mark up.

Freedom isn't free. Unless you want to live in a world were your children have to read the bible in class sacrifice has to be made.

Paying a premium for inferior service quality is not exactly freedom

Re: Hetzner is working on LLM Inference

#80

Post by yjftsjthsd-h »

atherton94027 wrote: post #72
swiftcoder wrote: post #62

Earlier quoted context omitted.

> don't run 7 year old discounted servers That's a pretty big part of why, honestly. How many of us are running server workloads that actually need the price/performance tradeoff of cutting edge server hardware?

We're not even talking cutting edge. If you browse hetzner's auction page you can see they're renting machines with almost decade old CPUs like the Xeon W-2145 or i7 6700. That's just a lot of old hardware..

Well. Yeah. The auction page is the leftovers; if you want current stuff then you're in the wrong spot on their site
Post reply on HN