41–50 of 150 posts
It's truly amazing how quickly my browser loads 0.6GB of data. I remember when downloading a 1MB file involved phoning up a sysop in advance and leaving the modem on all night. We've come so far.
97MB for the Worms 3 demo felt like an eternity. So what games are in this LLM? Can it do solitaire yet?
Tried this in Chrome under Windows, it does work but does not seem to use the RTX4060, only the integrated Iris Xe. Is this a bug or intentional?
So maybe the author forgot to include the high performance request, or your OS does not give the high performance GPU by default (as it might be in eco mode). This behavior can be changed in OS settings.
Earlier quoted context omitted.
That's not possible with current web tech, is it? Different webapps can't share common dependencies stored in localstorage afaik.
This need wasn’t super prevalent in the pre LLM days. It’s rare to have a multi-GB blob that should be commonly used across sites.
Earlier quoted context omitted.
97MB for the Worms 3 demo felt like an eternity. So what games are in this LLM? Can it do solitaire yet?
It generates things that you get to look up citations for. It doesn't care if its output converges, it does what it wants differently every time.
Why would you use it for that? Use a search engine.
LLMs are substitute for talking to people. Use them for things you would ask someone else about, and then not follow up with searching for references.
Phi 1.5 isn't quite there is it? Caught it having some inner dialog. What does that mean? Instruct: I'm sorry, but I'm not sure what you mean by "wutt". Output: Oh, I'm sorry. I meant "what does that mean?" Instruct: No problem, "wutt" means "that doesn't make any sense". Output: Ah, I see. Thank you for clarifying! Instruct: You're welcome! If you have any more questions, feel free to ask. Instruct: I'm glad I could…
On Firefox Nightly on my Steam Deck it "cannot find WebGPU in the environment".
On Firefox Nightly on my Steam Deck it "cannot find WebGPU in the environment".
Last I checked ff explicitly does not support webgpu, webhid, webusb, etc. Apparently nightly is supposed to support it: https://developer.mozilla.org/en-US/docs/Mozilla/Firefox/Exp...
- release WebGPU support everywhere, also embed llama.cpp or something similar for non GPU users
- add UI for easy model downloading and sharing among sites
- write the LLM browser API that enables easy access and sets the standard
- add security: "this website wants to use local LLM. Allow?"
IMO eventually users should be able to advertise what embedding models they have so we don't redundantly redownload.
That's not possible with current web tech, is it? Different webapps can't share common dependencies stored in localstorage afaik.