Anything that reduces the footprint of LLM’s is welcome, however…

• making LLM compute cheaper in datacentres won’t mean lower total power/cooling/space/water consumption, like adding lanes and traffic, it will just mean more usage as it gets cheaper (and a short-term bump in margins for the LLM owners)
• these are still highly dedicated chips that are always going to be bound up in the mega-scale datacentre deployments
  • what happens if there is a paradigm shift in the exact compute architecture? Loads of junk servers and no applications able to make use of such a glut
  • these do nothing to push LLMs out of the datacenter and into non-corporate hands, which is the only spot where we might see fewer privacy concerns, less corporate control etc

If we’re stuck with the current compute/corporate paradigms, at least alternatives nibling at the unhealthy dominance of nVidia and the cloud giants is some small benefit.

This is looking like a really fantastic feature. Even with a remappable keyboard, there’s some limit to the number of layers and modifiers you can juggle without a GUI and this will be available across any/all keyboards used on the Plasma device.

The trailing slash is just following cp’s own special treatment

# before
dir/
    file1
target/
    orig.txt



# no trailing slash - copy *dir* to target
cp -r /path/to/some/dir /my/target

# after
target/
    orig.txt
    dir/            # dir copied to target
        file1



# with trailing slash - copy *contents* of dir to target
cp -r /path/to/some/dir/ /my/target

# after
target/
    orig.txt
    file1            # contents of dir