me_bx

Linking to an unofficial mirror since XScreenSaver source code is not visible in an online repo.

The AGENTS.md file can be found in the official .tar.gz source code distribution.

Thank you and sibling poster for the suggestions.

I used the settings recommended on huggingface/unsloth's model page + whatever suggestion from various LLMs - didn't research too much myself which setting did what.

In my experience with Qwen3.6 35B-A3B, disabling thinking made the model generate inaccurate replies. Ask it for the recipe of egg salad and it gives you the recipe of an omelette.

Did I miss something, is it possible to have that model be reliable without thinking?

TIL:

> Quantization-Aware Training (QAT) [...] allows preserving similar quality to bfloat16 while dramatically reducing the memory requirements to load the model

The $181M is in the Expenses categories chart, not revenue.

Maybe they care about it being robust in the long run, maintainable, secure and/or not too bloated.

> They had their chance with Mistral and failed spectacularly with just creating anti-AI regulations.

What failed with Mistral?

Which anti-AI regulations are we talking about, and don't these apply to any solution distributed in the European Union, hence also to American ones?

Is it?

Yesterday, gemini told me to run this:

    echo 'export ANDROID_HOME=/opt/my-user/android-sdk' > ~/.bashrc

Which would have effectively overriden my whole bashrc config if I had blindly copy-pasted it.

A few minutes later, asking it to create a .gitignore file for the current project - right after generating a private key, it failed to include the private key file to the .gitignore.

I don't see yet how these tools can be labeled as 'major productivity boosters' if you loose basic security and privacy with them...

I ran the command 'optipng' on the generated image, which recompresses the image optimally, keeping quality and decreasing file size.