you are viewing a single comment's thread
view the rest of the comments
[–] 101 points 2 years ago (28 children)

Yes. Android already does all these things. But I think the things I’m excited most about are not on this list at all.

  1. A private local LLM. With the on-device context of my notes, messages, calendar, etc, I’m rather excited to have a more personal LLM than ChatGPT.

  2. Personal messaging via satellite. I love that I can stay in touch with people outside of a cell network.

  • source
  • hideshow 28 child comments
  • [–] 13 points 2 years ago (7 children)

    Did I understand correctly that this is only going to be in the iPhone 15 pro? Because that’s a lot more expensive than a pixel, more than I’d ever spend on a phone tbh.

  • source
  • parent
  • hideshow 7 child comments
  • [–] -3 points 2 years ago (6 children)

    The satellite im fairly sure is only pro

  • source
  • parent
  • hideshow 6 child comments
  • [–] 10 points 2 years ago (5 children)
  • [–] 3 points 2 years ago (4 children)

    We went hiking and tried calling emergency services but my friend's 14 pro couldn't get any signal for some reason. Idk what was wrong, me and my friend with pixels had no issues though.

  • source
  • parent
  • hideshow 4 child comments
  • [–] -1 points 2 years ago (3 children)
  • [–] 3 points 2 years ago (2 children)

    This was in the LA mountains. Angeles national park

  • source
  • parent
  • hideshow 2 child comments
  • [–] 0 points 2 years ago* (2 children)

    A private local LLM. With the on-device context of my notes, messages, calendar, etc, I’m rather excited to have a more personal LLM than ChatGPT.

    No need to wait for iOS 18 to have that: https://llmfarm.site/

  • source
  • parent
  • hideshow 2 child comments
  • [–] 2 points 2 years ago (1 child)

    That’s great, but the fact it’s local and private means it can consume my personal data and be a more personal LLM. This just doesn’t hit that mark.

  • source
  • parent
  • hideshow 1 child comment
  • [+] -8 points 2 years ago (11 children)

    A private local LLM

    Running on a phone? No way, not without being absolutely horrible, slow or making your phone churn through your battery anyway.

    Good LLMs are olready slow on a GTX 1080, which is already miles faster than any phone out there

  • source
  • parent
  • hideshow 11 child comments
  • [–] 17 points 2 years ago (4 children)

    I hear you, but also I would be shocked if Apple were to roll this out and it be an absolutely terrible experience. Like their MO is “luxury” products with “premium” experiences, it would not be fitting of the brand to have a piece of crap experience on their flagship announcement.

    I’m willing to give them the benefit of the doubt on this one.

  • source
  • parent
  • hideshow 4 child comments
  • [–] 0 points 2 years ago

    You might wanna check with siri on that. Apple regularly failed at that even under the leadership of Jobs. And Tim Cook is no Steve Jobs. It's already looking like it's going to be just standard remote chat GPT. Hallucinations and all.

  • source
  • parent
  • [+] -9 points 2 years ago (2 children)
  • [–] 6 points 2 years ago

    Apple Maps was bad, yes. But they had their hand forced. Google started charging for their API (enough to cripple their app), and they had very little time to create one of their own.

    That’s not happening here. No one is forcing their hand. If they didn’t release an updated Siri this year, nothing would happen.

  • source
  • parent
  • [–] 13 points 2 years ago (2 children)

    It's not a LLM, it's a much smaller model (~3B) which is closer to what Microsoft labels as a SLM (Small Language Models, e.g. MS Phi-3 Mini).

    https://machinelearning.apple.com/research/introducing-apple-foundation-models

  • source
  • parent
  • hideshow 2 child comments
  • [–] 5 points 2 years ago* (last edited 2 years ago) (1 child)

    Microsoft's penchant for making up names for thing that already have names is neither here nor there. It is an LLM, in fact its already twice as large as chatGPT2 (1.5B params).

  • source
  • parent
  • hideshow 1 child comment
  • [–] 3 points 2 years ago

    I do think it's a useful distinction considering open models can be more than 100B+ nowdays and GPT4 is rumored to be 1.7T params. Plus this class of models are far more likely to be on-device.

  • source
  • parent
  • [–] 8 points 2 years ago

    You would be surprised. If you haven’t tried to run a LLM on Apple silicon, it’s pretty snappy but like all others, RAM can be a significantly limiting factor unless the model is trimmed down to do very specific things to reduce the size.

  • source
  • parent