Hacker News: johnfahey

New comment by johnfahey in "Grok 4.7"

johnfahey — Mon, 21 Sep 2026 17:38:52 +0000

No doubt xAI has seen rapid progress, but it's been several months of them being "just behind" OpenAI and Anthropic. It seems the gap between just behind the frontier and pushing it is a lot wider than most people thought it was a year ago, and that's why a clear third contender in the frontier model space has yet to materialize.

Show HN: Nully – FOSS AI chat without the bloat

johnfahey — Wed, 08 Jul 2026 20:25:47 +0000

As someone who doesn't use any of the more "advanced" features on sites like ChatGPT or Claude (agentic mode, memory, image generation, deep research, etc), I found the bloat of these services, both in UX and in performance, to be pretty tedious.

There are a million AI chat interfaces out there, but I could never find one that just did simple messaging and history in the most minimal and lightweight way possible. So I made my own.

Nully is written in Go and Vanilla HTML/CSS/JS. It literally just sends messages to OpenRouter, receives the result, and stores your messages locally. It's very simple, very clean, and very lightweight (see performance graphs on the site).

It has basic chat features, basic model settings, and works with just about any OpenRouter text model. It supports attachments and basic web search via the OpenRouter API.

I made Nully mostly for myself, but the cost to open-source it and host it on a cheap VPS is practically nothing, so I figured I'd release it if it can help anyone else. I don't make any revenue off of it, and have no desire to. There's no accounts, no subscriptions, and no proxies or middlemen (other than OpenRouter of course). You can export all of your chats to JSON, and everything stays on your local machine. It's 100% open source and self-hostable if you want that too. Let me know what you think!

Comments URL: https://news.ycombinator.com/item?id=48836996

Points: 3

# Comments: 0

New comment by johnfahey in "Claude Sonnet 5"

johnfahey — Tue, 30 Jun 2026 18:13:50 +0000

Judging from those cost-performance graphs, Sonnet doesn't make sense to run at anything higher than a medium reasoning level, since Opus 4.8 low reasoning outclasses it for the price.

This line as a selling point is also pretty funny:

> Evaluations also show that it has a much lower ability to perform cybersecurity tasks than our current Opus models.