Mike Larkin

Low-level developer. Peakbagger. Private Pilot.

Founder/CTO Ringcube (acquired by Citrix) and Deepfactor (acquired by Cisco).

Building hypervisors and OSes.

[OpenBSD @openbsd]

Errata patches released for vmd, xorg, and OpenSMTPD: https://t.co/YTWc6G2KWy

[セyphrロiot @CyphrRiot]

Music done right... or is it riot?

https://t.co/6Dv3Osnhmx

[Mike Larkin @mlarkin2012]

@Arm A new era of mountaineering.

27.98825, 86.92514

[Mike Larkin @mlarkin2012]

Curse of McSorley's stick continues

[Georgi Gerganov @ggerganov]

llama.cpp now has an official website: https://t.co/vztdUpdBWL

Our goal is to make local AI accessible to everyone, and improving the user experience is a big part of that. On the new landing page you’ll find a single-line cross-platform installer. The installation provides a single unified `llama` entrypoint which you can use to run/serve models and interface with 3rd-party agentic applications.

While oriented towards simplified user experience, the new `llama` application also provides all the advanced functionality of the existing llama.cpp tooling with which experienced users are already familiar. Also note that all GGUF models that you might have already downloaded with llama.cpp in the past will be automatically available to use without downloading again (they are stored in the common HF cache on your machine).

We have many improvements in the pipeline both at the UX and at the engine level and we plan to iteratively ship new things over the coming months. One of the main focuses will be seamless integration with local-friendly 3rd-party agents (such as Pi). In the meantime, we’ll continue to listen for feedback from the community and adjust accordingly, so keep letting us know what you think and need.

[Ivan Fioravanti ᯅ @ivanfioravanti]

He started playing with Step 3.7 flash! Be ready! 🔥

[Mike Larkin @mlarkin2012]

@AlexFengzh @MrPeterLMorris nice. my nas only has PCIe x4 which means I'm not getting 50Gb/s even if the storage fabric could go that fast (and it can't, anyway).

[Mike Larkin @mlarkin2012]

@nicboul It's the firefly RISC-V Spacemit K3 machine.

[Mike Larkin @mlarkin2012]

Just received a box full of AI!

[Mike Larkin @mlarkin2012]

@MrPeterLMorris And in case it wasn't obvious, if you have more than one spark, they can all share the same HF cache from the NAS.

[Mike Larkin @mlarkin2012]

@hikingtheglobe0 Omaha, Nebraska

[セyphrロiot @CyphrRiot]

OpenRiot v7.9.x -- the OpenBSD desktop you've always imaged. Start a Riot at https://t.co/tAXPwGj6UC

[Mike Larkin @mlarkin2012]

@PhantomStays 10x engineers always get more done by staring at the back of their laptop instead of sitting in front of it.

[mr-r0b0t @mr_r0b0t]

This blew up in the best of ways! We've got @NVIDIAAI GB10 users of all skill levels, helping each other build and grow. Whether you're looking for help fine tuning or setting up your first GB10, you will inevitably find someone who is happy to help!
https://t.co/Z2m5vyYLLs

[Yoshihiro Kawamata @yoshi_kaw]

最近ちょくちょく見かける OpenRiot
https://t.co/NH02sbMn6z

「openBSD 7.9 用の、驚くほど洗練された i3 デスクトップ環境」だそうだ。

河豚板 7.9の上に載るかな。試してみよう
「ディスク 25GB以上推奨 100GB以上が最適」ってのが気になる。別パーティション切ったほうがいいかも

[Mike Larkin @mlarkin2012]

@EM_Creates @Dodgers no

[セyphrロiot @CyphrRiot]

“Why can’t an OpenBSD installation just work correctly and be usable without a hundred hours of fiddling?”

Now it can. Start a Riot https://t.co/Jbm3fbsLj4

There's an invite-only Signal group for issues as well as https://t.co/S84NI0BYow

WARNING: Very BETA.

[Mike Larkin @mlarkin2012]

@yoshi_kaw @x61sh RTL8821CE は間もなく利用可能になる可能性があります。この度は urtwx(4) と非常に類似しているため、近くコミットする予定だからです。

[Mike Larkin @mlarkin2012]

@Hoainguyen888 They made a TV appearance long before that one. This one's from 1990 and its pretty crazy. https://t.co/BEDDtwC3ik