Hacker News: linolevan

New comment by linolevan in "April 2026 TLDR Setup for Ollama and Gemma 4 26B on a Mac mini"

linolevan — Fri, 03 Apr 2026 16:01:57 +0000

What I really don't get is why more people don't talk about LMStudio, I switched to it months ago and it seems like a straight upgrade.

New comment by linolevan in "Google releases Gemma 4 open models"

linolevan — Thu, 02 Apr 2026 16:53:27 +0000

Hosted on Parasail + Google (both for free, as of now) themselves, probably would give those a shot

New comment by linolevan in "Qwen3.6-Plus: Towards real world agents"

linolevan — Thu, 02 Apr 2026 15:30:36 +0000

I’m surprised that people are surprised. Qwen has been hosting private plus and max variants for a while now.

New comment by linolevan in "Subscription bombing and how to mitigate it"

linolevan — Thu, 02 Apr 2026 04:54:21 +0000

Well written piece on an attack vector I'd never thought too hard about before. Thanks for elaborating on why sending an email or two to a random person helps an attacker achieve their goal. A lot of similar articles skip over details like that.

Trinity-Large-Thinking: Scaling an Open Source Frontier Agent

linolevan — Wed, 01 Apr 2026 19:13:13 +0000

Article URL: https://www.arcee.ai/blog/trinity-large-thinking

Comments URL: https://news.ycombinator.com/item?id=47605189

Points: 4

# Comments: 0

New comment by linolevan in "Reports of code's death are greatly exaggerated"

linolevan — Mon, 23 Mar 2026 01:28:54 +0000

How much is OpenAI paying you for this

New comment by linolevan in "Show HN: We built a terminal-only Bluesky / AT Proto client written in Fortran"

linolevan — Fri, 20 Mar 2026 23:57:09 +0000

https://tangled.org/ <--- GitHub on ATProto

that's all I'm aware of

(edit) Oops, just saw that you mentioned it, confused by your first line then. Tangled is awesome!

New comment by linolevan in "Marvin Hagemeister, Luca Casonato, David Sherret and Phil Hawksworth left Deno"

linolevan — Tue, 17 Mar 2026 22:21:37 +0000

John too:

https://bsky.app/profile/johndonmoyer.com/post/3mhbkepa7gs2v

New comment by linolevan in "Starlink Mini as a failover"

linolevan — Mon, 16 Mar 2026 21:04:46 +0000

Where on earth are you living with that kind of price point? Unreal.

New comment by linolevan in "A Recursive Algorithm to Render Signed Distance Fields"

linolevan — Sat, 14 Mar 2026 21:45:45 +0000

Played around with the code to implement a little bit of SIMD. Was able to squeeze out a decent improvement, ~250 fps avg, ~140 low, ~333 high (on an m4). Looks pretty straightforward to do threading with as well. Cool stuff! Could work to bring more gpu stuff back down to the cpu.

New comment by linolevan in "Claude Doubles Usage Limits During Off-Peak Hours (March 13–27, 2026)"

linolevan — Sat, 14 Mar 2026 21:10:06 +0000

Oops! Looks like we posted at the same time.

New comment by linolevan in "Claude March 2026 usage promotion"

linolevan — Sat, 14 Mar 2026 21:09:23 +0000

> Does bonus usage count against my weekly usage limit?

> No. The additional usage you get during off-peak hours doesn’t count toward any weekly usage limits on your plan.

New comment by linolevan in "Shipping Grayscale Photos at Small Scale"

linolevan — Sat, 14 Mar 2026 04:41:29 +0000

Did… you copy paste this from another discussion? I’ve read this comment before.

Why Can't Programmers Program? (2007)

linolevan — Fri, 13 Mar 2026 16:27:43 +0000

Article URL: https://blog.codinghorror.com/why-cant-programmers-program/

Comments URL: https://news.ycombinator.com/item?id=47366529

Points: 2

# Comments: 0

New comment by linolevan in "Launch HN: IonRouter (YC W26) – High-throughput, low-cost inference"

linolevan — Thu, 12 Mar 2026 23:45:01 +0000

According to the providers that I keep track of, Cumulus is typically pretty price competitive, except for MiniMax where DeepInfra and Together are much cheaper and GLM-5 where DeepInfra and z.AI's own hosting is much cheaper.

(Also technically qwen3 8b w/ novita being first place but barely)

New comment by linolevan in "Launch HN: IonRouter (YC W26) – High-throughput, low-cost inference"

linolevan — Thu, 12 Mar 2026 23:38:23 +0000

Can we get context length / output length docs (looks like you mention "Max tokens (chat)" of 128k but it's unclear what that means)? Also it looks like your docs page is out of date from your playground page.

Also piece of feedback: it kind of sucks to have glm/minimax/kimi on separate api endpoints. I assume it's a game you play to get lower latency on routing for popular models but from a consumer perspective it's not great.

Can LLMs Be Computers?

linolevan — Thu, 12 Mar 2026 20:02:23 +0000

Article URL: https://www.percepta.ai/blog/can-llms-be-computers

Comments URL: https://news.ycombinator.com/item?id=47356274

Points: 9

# Comments: 0

Claude Code Voice Mode

linolevan — Thu, 12 Mar 2026 17:18:31 +0000

Article URL: https://support.claude.com/en/articles/11101966-using-voice-mode

Comments URL: https://news.ycombinator.com/item?id=47354164

Points: 24

# Comments: 13

New comment by linolevan in "Sarvam 105B, the first competitive Indian open source LLM"

linolevan — Sat, 07 Mar 2026 21:37:25 +0000

Looks like at least a second release, they had one other LLM before this.

New comment by linolevan in "Workers who love ‘synergizing paradigms’ might be bad at their jobs"

linolevan — Fri, 06 Mar 2026 16:20:57 +0000

This is awesome. Almost all of these are believable even if you're looking at pretty carefully. I need this on a firestick or something.