Is GPT-6 50% cheaper?

> GPT‑6 Luna vs. GPT‑5.6 Luna | $0.20 → $0.10 | $1.20 → $0.50 | 50% cheaper

I can read it as follows (below), meaning that GPT-5.6 is 50% cheaper.

- GPT-6 = $0.20

- GPT-5.6 = $0.10

The table on https://developers.openai.com/api/docs/pricing is more readable:

  +--------------+-------+--------------+--------------+--------+
  | Model        | Input | Cached input | Cache writes | Output |
  +--------------+-------+--------------+--------------+--------+
  | gpt-6-luna   | $0.10 | $0.01        | $0.125       | $0.50  |
  | gpt-5.6-luna | $0.20 | $0.02        | $0.25        | $1.20  |
  +--------------+-------+--------------+--------------+--------+

Yeah, how they put, is confusing to me, they should have put that table instead of what they have right now in the article.

Yes, GPT-6 Luna is 50%-58% cheaper than GPT-5.6 Luna. (I think the blog text and graphs make it pretty clear.)

Yeah, but it confuses me, I read left to right, so if they put GPT-6 and $0.20 first, I would assume that's the new pricing, they should make it clear, not confusing.

I would think about if it was at least Go or Rust, but yet another interpreted language built on top of other abstractions? No thanks, I had my fair share of Xamarin, React Native, Electron, Tauri and TBH building native is the best (more work, sure, but at least when you have bugs or issues, it's easier to find solutions), with abstractions you need to deal with bugs on each layer.

Also in the era of AI, sharing native code is easier than ever, just compile down to WASM for the web and for the other binary.

And different price tag

Not only Lisp, but Javascript and Ruby (AKA dynamic languages) devs also rely heavily in REPLs, for me at least the reason is that you don't know what value a variable holts, so you need to inspect them somehow, with typed languages you have less reasons to reach such tool.

iPython was famous for a while but I think its popularity has waned, but yes as you point out really any dynamic language can do a repl; it's just that it's mostly the lisp/scheme and forth families that have leaned so heavily into the idea.

Not only dynamic languages! Java has `jshell` for instance

I got to the conclusion that boring is good enough, for me NPM is the best tool, is not that fast, but the installation speed is not that bad, considering how many times I need to run from scratch, the difference does not matter, it is stable and it's the default Node package manager, so I don't need to bother to install in CI any other tool, etc. My experience for long living projects is to stick with the defaults, because in a few years third party tools will change, it happen to my team with Yarn classic, now that we are with vanilla Node and NPM no worries, same with 3rd party dependencies, just use as less as possible.

Unless you care about security, stability, DX, speed and correctness then npm is fine choice.

Describing NPM as boring is a stretch. Given its security model, i think spicy is a far more apt label.

I have switched to pnpm already.

Npm likes to run package lifecycle scripts of your dependencies by default. You have to explicitly opt out of this.

It’s probably the worst package manager out of npm, pnpm, yarn, and bun.

I have the same experience - npm was bad, now it is fast enough.

yeah the problem is npm is now Microsoft, just like Github. I trust Node. I don't trust NPM.

So weird, launching M6 so silently.

Allegedly, they aren't making a big deal out of the M6 which won't have Pro, Max or Ultra configs at all, instead waiting for the M7: https://9to5mac.com/2026/08/08/apple-m7-chip-heres-why-it-ma...

Interesting, thanks for sharing.

Wasn’t their newer text editor/IDE written in Rust? I forgot how it was called and also didn’t followed up with their development.

It was called Fleet (I almost forgot about it too) and has since been retired: https://blog.jetbrains.com/fleet/2025/12/the-future-of-fleet...

(if you wanted to be snarky, you could say that Fleet dissolved into thin Air)

No, Fleet was written in Kotlin running on the JVM. I seem to remember it used Skia for the UI.

In 2018 when Windows and macOS used 2GB just after boot, I tried Arch Linux and with a neat window manager (dwm) a clean boot was using just 300MB, that means our OS is wasting just too much RAM for nothing. This year with Fedora Atomic 44 + Niri is using 800MB, so I can give a second life to my old device from 2013.

Did you check if the resources were consumed by the LSP or Zed? Also there was a Zed fork without the AI stuff, not sure if is a hard fork or not and if is still maintained (IIRC wasnt hosted on Github)

Gram Editor. It's actively maintained.

I know people disagree over definitions of "hard fork". Gram does not send patches to Zed but does occasionally merge fixes from Zed. There is not any specific effort to keep (non-AI) feature parity.

https://gram-editor.com

https://codeberg.org/GramEditor/gram

Why does the author hate the AI so much?

have they fixed the file refresh issue?

For me, what the browser needs it's a good DX to build components, I kind of like Svelte and Vue in that regard, where your CSS, HTML and JS lives all together.

Writing a CSS file with the time will start growing and at some point will be very hard to know what to keep and what to delete, I think this is the reason why tailwind became popular, because it keep only the used rules, now the downside is that now your html elements has ton of classes.