Hacker News: jonas_scholzhttps://news.ycombinator.com/user?id=jonas_scholzHacker News RSShttps://hnrss.org/hnrss v2.1.1Mon, 21 Sep 2026 10:26:55 +0000<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>They wont, they are very public about never subsidising any products. I doubt they change that after 29 years now.

]]>Sat, 25 Jul 2026 15:32:55 +0000https://news.ycombinator.com/item?id=49048449jonas_scholzhttps://news.ycombinator.com/item?id=49048449https://news.ycombinator.com/item?id=49048449<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>also slightly disrespectful to think that this was a single prompt, but ok:D

]]>Fri, 24 Jul 2026 19:42:09 +0000https://news.ycombinator.com/item?id=49040695jonas_scholzhttps://news.ycombinator.com/item?id=49040695https://news.ycombinator.com/item?id=49040695<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>yeah, but why would LLM inference now triple the price of the other products? Or am I misunderstanding what you're saying?

]]>Fri, 24 Jul 2026 17:36:55 +0000https://news.ycombinator.com/item?id=49039105jonas_scholzhttps://news.ycombinator.com/item?id=49039105https://news.ycombinator.com/item?id=49039105<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>"no reason" like hardware prices going through the roof?

]]>Fri, 24 Jul 2026 15:22:55 +0000https://news.ycombinator.com/item?id=49037013jonas_scholzhttps://news.ycombinator.com/item?id=49037013https://news.ycombinator.com/item?id=49037013<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>Why do you think this would be madness? It seems like, at least in the EU, there is barely competition for the big open-weight models?

]]>Fri, 24 Jul 2026 12:46:30 +0000https://news.ycombinator.com/item?id=49034799jonas_scholzhttps://news.ycombinator.com/item?id=49034799https://news.ycombinator.com/item?id=49034799<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>fine, i agree that this doesnt sound like me. But the content is correct and not a hallucination!

]]>Fri, 24 Jul 2026 12:37:22 +0000https://news.ycombinator.com/item?id=49034702jonas_scholzhttps://news.ycombinator.com/item?id=49034702https://news.ycombinator.com/item?id=49034702<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>I mostly use Hetzner baremetal servers and not cloud, but the support there is in my experience very competent, but also VERY german and direct. I can imagine if youre neither german or super technical that this can be intimidating or considered unfriendly

]]>Fri, 24 Jul 2026 12:27:00 +0000https://news.ycombinator.com/item?id=49034600jonas_scholzhttps://news.ycombinator.com/item?id=49034600https://news.ycombinator.com/item?id=49034600<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>what is the hallucination here? it is fast, free and fun to try. And I genuinely think that the hardware decision (if they get bigger gpus) decides if this will be a banger product or not?

]]>Fri, 24 Jul 2026 12:20:47 +0000https://news.ycombinator.com/item?id=49034527jonas_scholzhttps://news.ycombinator.com/item?id=49034527https://news.ycombinator.com/item?id=49034527<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>yea, but no prompt caching right? This makes it unusable for my usecase at least, the cost would be insane

]]>Fri, 24 Jul 2026 11:45:59 +0000https://news.ycombinator.com/item?id=49034210jonas_scholzhttps://news.ycombinator.com/item?id=49034210https://news.ycombinator.com/item?id=49034210<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>I really hope they dont stop at the small models though! The bigger ones that dont fit on a single GPU are more interesting I think

]]>Fri, 24 Jul 2026 11:40:25 +0000https://news.ycombinator.com/item?id=49034150jonas_scholzhttps://news.ycombinator.com/item?id=49034150https://news.ycombinator.com/item?id=49034150<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>agree. I have a usecase for a bigger open-weight model hosted by an EU company and the selection isnt really great, hard to make the regulatory gods (and the devs) happy at the same time right now

]]>Fri, 24 Jul 2026 11:35:02 +0000https://news.ycombinator.com/item?id=49034102jonas_scholzhttps://news.ycombinator.com/item?id=49034102https://news.ycombinator.com/item?id=49034102<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>My writing wasnt clear here I think, without the option it defaults to reasoning enabled. With "without it" I meant without enable_thinking=False!

]]>Fri, 24 Jul 2026 11:09:22 +0000https://news.ycombinator.com/item?id=49033870jonas_scholzhttps://news.ycombinator.com/item?id=49033870https://news.ycombinator.com/item?id=49033870<![CDATA[New comment by jonas_scholz in "Hetzner is working on LLM Inference"]]>Interesting, didn’t know about them! Weird model selection though, no glm or deepseek?

]]>Fri, 24 Jul 2026 10:37:17 +0000https://news.ycombinator.com/item?id=49033584jonas_scholzhttps://news.ycombinator.com/item?id=49033584https://news.ycombinator.com/item?id=49033584<![CDATA[Hetzner is working on LLM Inference]]>Article URL: https://sliplane.io/blog/hetzner-inference

Comments URL: https://news.ycombinator.com/item?id=49033087

Points: 156

# Comments: 86

]]>Fri, 24 Jul 2026 09:24:44 +0000https://sliplane.io/blog/hetzner-inferencejonas_scholzhttps://news.ycombinator.com/item?id=49033087https://news.ycombinator.com/item?id=49033087<![CDATA[We built our own DNS server]]>Article URL: https://sliplane.io/blog/how-we-built-our-own-dns-server

Comments URL: https://news.ycombinator.com/item?id=47817068

Points: 2

# Comments: 0

]]>Sat, 18 Apr 2026 16:19:39 +0000https://sliplane.io/blog/how-we-built-our-own-dns-serverjonas_scholzhttps://news.ycombinator.com/item?id=47817068https://news.ycombinator.com/item?id=47817068<![CDATA[New comment by jonas_scholz in "Netbird – Open Source Zero Trust Networking"]]>we love and support netbird at sliplane <3 https://docs.sliplane.io/private-networking/netbird

]]>Mon, 02 Feb 2026 16:04:58 +0000https://news.ycombinator.com/item?id=46857549jonas_scholzhttps://news.ycombinator.com/item?id=46857549https://news.ycombinator.com/item?id=46857549<![CDATA[What the Fuck Is a Vcpu?]]>Article URL: https://sliplane.io/blog/vcpus-are-a-marketing-scam

Comments URL: https://news.ycombinator.com/item?id=46753797

Points: 2

# Comments: 0

]]>Sun, 25 Jan 2026 13:07:39 +0000https://sliplane.io/blog/vcpus-are-a-marketing-scamjonas_scholzhttps://news.ycombinator.com/item?id=46753797https://news.ycombinator.com/item?id=46753797<![CDATA[New comment by jonas_scholz in "Show HN: European tech news in 6 languages"]]>Thank you for mentioning Sliplane! Really cool project.

]]>Sat, 15 Nov 2025 19:26:32 +0000https://news.ycombinator.com/item?id=45939919jonas_scholzhttps://news.ycombinator.com/item?id=45939919https://news.ycombinator.com/item?id=45939919<![CDATA[Fly Away from Fly.io]]>Article URL: https://sliplane.io/blog/fly-away-from-fly

Comments URL: https://news.ycombinator.com/item?id=44193267

Points: 4

# Comments: 0

]]>Thu, 05 Jun 2025 16:37:40 +0000https://sliplane.io/blog/fly-away-from-flyjonas_scholzhttps://news.ycombinator.com/item?id=44193267https://news.ycombinator.com/item?id=44193267<![CDATA[New comment by jonas_scholz in "Firecracker vs. QEMU"]]>afaik no firecracker support with virt-manager. There's firectl to make it a bit easier but the tooling is generally still pretty rough

]]>Thu, 06 Mar 2025 14:15:03 +0000https://news.ycombinator.com/item?id=43280424jonas_scholzhttps://news.ycombinator.com/item?id=43280424https://news.ycombinator.com/item?id=43280424