Also a speech to speech model it can't do any tool use or you can't guide it much am I mistaken? How will it be used for agentic uses is the part I don't get. Edit: this is response from Argon if anyone was also curious: You are not mistaken about Dessa: because it was trained solely on raw audio with zero text data, it cannot read text system prompts or output code for tool calls on its own. To use it as an agent, developers must either steer it purely through spoken voice instructions or pair it as a conversational "voice front-end" alongside a separate text LLM that listens to the audio and executes the tools in the background.
In Austin, I've never been in one but I constantly see them making weird decisions and once, it was a red on left turn a waymo ahead of me and me behind it the turn left turned green it didnt move light turned red, next time it turned green it moved. I really think they're safe cause they have learned to do what others do in front of them and when they have to make their own decisions everything falls apart. I do hope it gets better a little faster.
Totally understand thanks for the explanation. still it doesnt make sense and answer why are they so strict about using cooling devices. If theyre using renewables and stuff it only makes sense cooling would be more okay that using that same energy to heat houses, yet cooling is just political and restricted. Im in US and i could care less it just reading comments even on here how hard can be installing AC and stuff really doesn't make sense.
https://x.com/WallStreetApes/status/2035921843757216158?s=20 This is just one example. EVgo and many other very large charging networks use generators also. Even if you charge at home or anywhere else which you don't see the generator, that electricity has to be made from using coal gas or other stuff, renewables are not a big share in most of them. Those cables transferring electricity loses a lot of energy and gives off lots of heat but you won't see that.
I seriously wasnt judging i maybe shouldve asked llm to frame it better cause i knew it might sound that way, thats why i added: im not judging, merely asking...
So basically because they just browsed and used the information that was mostly public on the internet and they didnt copy it, they just learned from it and thats fine. Which makes sense. None of the llms let u copy someones work exactly anyways... makes total sense honestly. So in this case what happens to distilling? Is that also learning or ur trying to get to their actual weights by kind of reverse engineering it? Where would the argument fall there?
Thanks to a Huggingface linked below, I tested it and im not impressed. prmopt: i need to contact my boss i will be late. Result: 20mins [{"name":"set_timer","arguments":{"time_human":"20 minutes"}}]. It didnt use the email tool and i tried 2-3 different ways of asking it.
instead of cursing and attacking, you can have actual discussions. most of the comments in this thread are hating towards trump and his policies without actually discussing any of the policies. Maybe you can't even read the comments. Just blank statements and making fun of trump and his policies. If u have any counters to them u will get flagged and downvoted to hell. Just like your comment, HN has gone from an actual intellectual discussions, to Bluesky style hatred and making fun. Good job proving my point.