m_kos

Sadly, I can't run GLM 5.2 locally.

I get that people here have reasons to hate on Ollama but it has two genuinely strong points: - it makes it very easy to use open models with a lot of harnesses and assistants via `ollama launch`; Their $20 subscription is very generous, and they claim not to log or train on your data.

I will be curious to see if their subscription ever supports the ~3T open weights models announced this week.

Nicely done! Would you consider adding an Agent SDK login for Claude for those who don't have an API budget and want to avoid potential issues with Anthropic's capricious policies around CLI use in third-party apps?

Rageprompting

Lovely name! I implemented profanity monitoring in my Hermes setup to identify "learning opportunities" for my agents. It is quite useful. If you are budget-conscious, one challenge is determining what is the smallest number of previous rounds that Hermes needs to correctly infer what it did wrong. Curiously, Claude Code is horrible at figuring out what it did wrong. I often read its memories, and they are rarely useful.

$300 is my employer's monthly cap on Claude Enterprise. It lasts me at most a week of moderate use. I would much rather get Codex Pro and Claude Pro or Max, which would cost ≤ $200. For $300, one could also add Gemini Ultra to the mix so I could have all three review each other's code, etc.

Claude can be very good but enterprise pricing doesn't make sense to me.

What price range, if you don't mind sharing?

Thanks. I will try this! I need to read up on how to work with vision models for both generation and understanding.

Very useful website. Would you have insight into what models are best at editing existing images?

I often have to make very specific edits while keeping the rest of the image intact and haven't yet found a good model. These are typically abstract images for experiments.

I asked gpt-image-2 to recolor specific scales of your Seedream 4 snake and change the shape of others. It did very poorly.

Neat! I might give it a try.

What do you mean by interfaces in "These interfaces can do literally anything on the host machine. You're responsible for your own security"?

Also, your backdooring image links to a 404.

Good luck!

There has been promising work on olfactory training, which you can do very inexpensively at home. If you can, I would consider seeing any ENT first to rule out polyps, etc.

I am not surprised you tried given your earlier project :) https://hackaday.com/2022/09/25/this-found-sound-organ-was-m...

Maybe at the bottom your marbles could land on surfaces with different accustic properties. Track selection would determine the surface and release time would determine the timing.

I live in a big city. DoorDash always delivers to an address 10 minutes away walking from my building. It is quite inconvenient in the winter. With human drivers, you can at least try to convince them to use Google Maps, but with an AI?

On second thought, prompt injection via delivery instructions?

For synth heads, there is also a well-regarded Syntorial app/course. (BTW, has anyone tried their Building Blocks? There are very few online reviews.)

What I don't quite get is why manufacturers of midi controllers (Arturia, Novation, NI, etc.), with the exception of, possibly only Korg, don't release any of their digital instruments as mobile apps. After sitting the whole day in front of my computer, the last thing I want to do is to swap VS Code for Ableton or Kontakt and spend a few more hours in the glow of my monitors.

(I do get that if you are very serious about making music you need a proper computer set up. I am just a mere amateur hobbyist.)

The m8 and the recently, heavily promoted Woovebox 2 are the Emacs/Vim of grooveboxes. They hide a vast amount of functionality behind what initially seems like an impenetrable jungle of button presses and shortcuts; a system that ultimately proves* to be highly ergonomic.

*Based on what I read. Sadly, I don't own these devices.

https://musictech.com/news/bitwig-studio-4-spectral-suite-ap...

If you have to use, e.g., CUDA Toolkit 11.8, then you need a specific version of VS and its build tools for CUDA's VS integration to work. I don't know why exactly that is and I wish I didn't have to deal with it.

Why is it so hard to install PyTorch, or CUDA, or libraries like FlashAttention or DeepSpeed that build against PyTorch and CUDA?

This is so true! On Windows (and WSL) it is also exacerbated by some packages requiring the use of compilers bundled with outdated Visual Studio versions, some of which are only available by manually crafting download paths. I can't wait for a better dev experience.

GitHub of the person who prepared the data. I am curious how much compute was needed for NY. I would love to do it for my metro but I suspect it is way beyond my budget.

https://github.com/yz3440

(The commenters below are right. It is the Maps API, not compute, that I should worry about. Using the free tier, it would have taken the author years to download all tiles. I wish I had their budget!)

I've never seen Claude Code do anything that git couldn't easily reverse, so I'm not that worried about that any more.

I had one agent that installed something with pip and used it to rewrite my git history. It used the wrong flag and what was left was one file. I am glad I caught it before auto-commit and push triggered. It was one of my hobby projects, so not a huge deal. I can't imagine the fallout of this happening in a work environment.

Well done! Very clean design and easy to use. Initially, I missed some conveniences of modern piano rolls (like tap and hold to area select a bunch of notes), but ultimately it just made me more careful, which was an interesting change. Some UX decisions were very thoughtful, like the blue vertical lines.

If I could suggest anything, it would be to let fat-fingered users like me hide the left menu in the landscape orientation (mobile) and stretch the roll sideways.

Thank you!

One component of my hobby web app project is a wavetable. Below are two examples of wavetables. I want it to not tax the browser so that other, latency sensitive, components do not suffer.

Would you have any suggestions on what JS/TS package to use? I built a quick prototype in three.js but I am neither a 3D person nor a web dev, so I would appreciate your advice.

Examples:

- https://audiolabs-erlangen.de/media/pages/resources/MIR/2024...

- https://images.squarespace-cdn.com/content/v1/5ee5aa63c3a410...

Good read – I have learned something new.

I have been having a horrible experience with Sonnet 4 via Cursor and Web. It keeps cutting corners and misreporting what it did. These are not hallucinations. Threatening it with deletion (inspired by Anthropic's report) only makes things worse.

It also pathologically lies about non-programming things. I tried reporting it but the mobile app says "Something went wrong. Please try again later." Very bizarre.

Am I the only person experiencing these issues? Many here seem to adore Claude.

GH repo linked below but the README is pretty minimal:

https://github.com/playbenny/benny

I think it is a great idea, but the images on your page and lack of pricing details make me think it is expensive and focused on "manly" interests (gambling, whiskey, etc.).

It is not my project.

[Tangent] Anyone here using 2.5 Pro in Gemini Advanced? I have been experiencing a ton of bugs, e.g.,:

- [codes] showing up instead of references,

- raw search tool output sliding across the screen,

- Gemini continusly answering questions asked two or more messages before but ignoring the most recent one (you need to ask Gemini an unrelated question for it to snap out of this bug for a few minutes),

- weird messages including text irrelevant to any of my chats with Gemini, like baseball,

- confusing its own replies with mine,

- not being able to run its own Python code due to some unsolvable formatting issue,

- timeouts, and more.

I saw an encoder similar to the one you like in an ultrasound once. Here is another: https://www.ctscorp.com/Resources/Press-Releases/New-Three-i...

There is something satisfying in noticing the same solution applied to problems in different domains, like audio, medicine, and aviation.