Hacker News: Abishek_Muthian

New comment by Abishek_Muthian in "The Age of Personalized Hardware Is Coming"

Abishek_Muthian — Mon, 06 Jul 2026 03:42:45 +0000

With hardware/computers becoming locked, thin clients and unaffordable we're already forced to customize our devices cyber punk style.

You're lucky if you're in a region where these open-hardware companies sell their wares, even though many of them will go under in the current market.

New comment by Abishek_Muthian in "Jamesob's guide to running SOTA LLMs locally"

Abishek_Muthian — Sat, 04 Jul 2026 04:02:37 +0000

Absolutely true. All this craze about running coding LLMs locally has been detrimental to local AI where purpose built SLMs could actually be beneficial.

Little tools for NLP, TTS, image processing, audio engineering, signal processing, diffusion plugin for Krita etc. are all great for local setup. I wrote a small piece on it few days back[1].

[1] https://abishekmuthian.com/multiple-20-ai-plans-are-better-t...

New comment by Abishek_Muthian in "SearXNG: A free internet metasearch engine"

Abishek_Muthian — Sat, 04 Jul 2026 03:45:59 +0000

This is great, like many others I've been thinking of something like hister but only for bookmarked web pages. I presume it should be straightforward with hister to do that?

All the best!

New comment by Abishek_Muthian in "PeerTube is a free, decentralized and federated video platform"

Abishek_Muthian — Fri, 03 Jul 2026 04:07:15 +0000

Interesting, I thought the major issue with PeerTube was the high latency due the decentralized nature and that they don't have the backing of Google infrastructure.

What you state sounds more like a chicken and egg problem. You'll publish content only when there are enough viewers but there can be viewers only when you publish content.

Like others have said, There's not much effort to publish in both platforms; good way to hedge against the algorithmic risks.

New comment by Abishek_Muthian in "Claude Sonnet 5"

Abishek_Muthian — Wed, 01 Jul 2026 15:29:50 +0000

4090 RTX laptop and a M4 Mac Mini, I've posted more information in my blog mentioned in another comment.

New comment by Abishek_Muthian in "Claude Sonnet 5"

Abishek_Muthian — Wed, 01 Jul 2026 15:29:07 +0000

I've posted in another comment above.

New comment by Abishek_Muthian in "Claude Sonnet 5"

Abishek_Muthian — Wed, 01 Jul 2026 15:28:14 +0000

Here's the blog - https://news.ycombinator.com/item?id=48748443 .

Multiple $20 AI Plans Are Better Than a Single $100 AI Plan

Abishek_Muthian — Wed, 01 Jul 2026 15:26:30 +0000

Article URL: https://abishekmuthian.com/multiple-20-ai-plans-are-better-than-a-single-100-ai-plan/

Comments URL: https://news.ycombinator.com/item?id=48748443

Points: 2

# Comments: 0

New comment by Abishek_Muthian in "Claude Sonnet 5"

Abishek_Muthian — Wed, 01 Jul 2026 04:15:09 +0000

I'm a senior skilled developer and I find Anthropic $20 + Open AI $20 + OpenCode Go $10 offers more value than $100 on any particular service.

Juggling between all different models/agents is quite simple with Zed.

A caution about OpenCode Go though, the entire company seems to be run by AI so there's lot of billing related issues with zero support. I subscribe new every month as I lost money due to double payment with automatic subscription.

For non coding related tasks I use local models.

P.S. If anyone is interested to read more about my setup, let me know I'll publish a blog post.

New comment by Abishek_Muthian in "Qwen 3.6 27B is the sweet spot for local development"

Abishek_Muthian — Tue, 30 Jun 2026 03:33:28 +0000

I agree completely. I think local AI is best limited to purpose built SLMs; all this craze around running quantized coding LLMs has taken the attention off SLMs.

New comment by Abishek_Muthian in "Qwen 3.6 27B is the sweet spot for local development"

Abishek_Muthian — Tue, 30 Jun 2026 03:24:19 +0000

>Sure you can use it in clamshell mode

Wouldn't this damage the MBP display?

My RTX laptop has air intake underneath the keyboard and clamshell mode is surely a recipe for disaster; I've taken numerous measures to ensure that the laptop doesn't stay awake when the lid is down.

New comment by Abishek_Muthian in "Show HN: DRM-Free Books"

Abishek_Muthian — Mon, 29 Jun 2026 03:32:35 +0000

libro.fm for DRM free audio books, buying from there supports local book shops.

New comment by Abishek_Muthian in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

Abishek_Muthian — Fri, 26 Jun 2026 15:18:48 +0000

True, can't beat current Apple chips on power efficiency.

New comment by Abishek_Muthian in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

Abishek_Muthian — Fri, 26 Jun 2026 07:30:30 +0000

>Whatever "Apple Tax" applies to a device that can run a capable model offline will amortise itself in a blink.

Current high-end Mac Studio with 32-core M3 Ultra chip and 96 GB of memory is $6800, 96GB is not enough to run GLM 5.2 without extreme quantization or stacking HW; but for the sake of discussion let's run quantized version on a single high end Mac Studio.

GLM 5.2 Max plan costs $ 112/m, so it would take ~60 months to recover the costs assuming the machine was bought just for AI. By then the current AI landscape would have changed drastically.

I use local AI on both Linux and Mac every single day, there's freedom, privacy and peace of mind in running the model locally. But I feel cost/value of local AI is overblown.

New comment by Abishek_Muthian in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

Abishek_Muthian — Fri, 26 Jun 2026 06:56:11 +0000

That's true only for the entry level macs. My M4 Mac Mini has the best Performance/value. But my workstation laptop with 32 cores, 96GB DDR5, Nvidia GPU costs lesser than Macs with lesser performance; not to mention I upgraded the RAM post purchase.

New comment by Abishek_Muthian in "Apple to skip high-end M6 Mac chips in favor of AI-focused M7 line"

Abishek_Muthian — Fri, 26 Jun 2026 04:00:25 +0000

It's plausible but is the Apple Tax for a 1TB memory machine on top of current memory prices really worth it? I paid around $4000 for 4090m laptop with 16GB VRAM back in 2023, it's great but DoA for even quantized LLMs. I can run SLMs and fine tune it but that's it.

We need one of those specialized inference chip startups to succeed and a PC manufacturer willing to bet on them against Nvidia for the local AI to find mass market appeal.

New comment by Abishek_Muthian in "Deno Desktop"

Abishek_Muthian — Mon, 22 Jun 2026 09:24:50 +0000

I agree about the complexity, but the core developers understand that the accessibility support is crucial and that's where my hope comes from.

[1] https://github.com/zed-industries/zed/discussions/6576

New comment by Abishek_Muthian in "Deno Desktop"

Abishek_Muthian — Mon, 22 Jun 2026 09:17:19 +0000

Interesting, I'm on Nvidia in Wayland most of the time too and haven't had single freeze or crash in a very long time. Even the Windows is running inside the Qemu.

What DE? I'm on KDE.

New comment by Abishek_Muthian in "Deno Desktop"

Abishek_Muthian — Mon, 22 Jun 2026 08:51:48 +0000

Sounds similar to Wails. It does the same but with Go instead of Rust.

New comment by Abishek_Muthian in "Deno Desktop"

Abishek_Muthian — Mon, 22 Jun 2026 06:40:04 +0000

Every time I use Zed across Linux, macOS and Windows , I'm amazed stable and performant it's GPUI framework is. As a user, I'm very happy with it; of course some basic features like accessibility is missing but I'm sure it will be implemented soon.

As a developer, I'm not sure what's the barrier for entry is apart from Rust then again it's the USP as well.