Intresting. But I’m curious about the performance.
A bigger LLM (mixtral) already struggles to run on my mid-range gaming PC. Trying to run an LLM that isn’t terrible on a standard laptop wouldn’t be a good experience.
I have no idea how this is set up to work technically, but most of the heavy lifting is gonna be on the GPU. I’m not sure that it matters much whether the browser is what’s pushing data to the GPU or some other package.
The thought that internet becomes shitty enough that you need a GPU to browse it is really frightening me. If we really reach that point that may be to run an AI which filters out AI generated spam which would really depress me 😭
The thought that internet becomes shitty enough that you need a GPU to browse it is really frightening me.
I mean, there was a point where an FPU was a separate chip and wasn’t the norm; now it’s built into the CPU.
I think that it’s probably safe to say that, in the future, there will be broader use of parallel processing, as we’ve fundamental limits on what we know we can do there with existing laws of physics with serial processing. That could wind up being part of the CPU. It could live on a separate piece of hardware – which may not necessarily be a “GPU” – parallel processing hardware entered the PC because the most-immediate need was to do 3d graphics rendering, but as you can see from the LLMs that people are running on GPUs today, that’s not the only application. The parallel compute accelerator cards that Nvidia is selling today for an arm and a leg on servers aren’t aimed at doing 3d graphics.
It may not be 3d graphics rendering or running LLMs that becomes the primary application. But I’d be reasonably comfortable saying that down the line, relative to today, there will be more parallel-processing hardware in computers than is present today.
For a lot of newer processors, it already is. Intel, Apple, Samsung, and Qualcomm like to brag about their processors having some sort of neural contraption meant to assist with AI processing.
If it stays around, it might be good enough that you don’t need a GPU to do it, since the CPU has an onboard chip that can handle that work instead, since tensor processors like that are a bit more efficient than GPUs, but are also more specialised.
Most people probably don’t have a dedicated GPU and an iGPU is probably not powerfull enough to run an LLM at decent speed. Also a decent model requires like 20GB of RAM which most people don’t have.
Not exactly. Most integrated chips have a small pool of dedicated VRAM, and then a bit more that they share with the system memory, though it’s generally only a portion, not all of it. It’s only Apple’s unified memory, and maybe other mobile chips that has them both share memory pool entirely, for better or worse, as far as I’m aware.
But it is worth noting that if you don’t have enough VRAM and have to put it into RAM, the minimum expectation is that you have twice the amount of RAM space. So if you have a GPU with 4GB of VRAM, and need to offload the extra to the system, you don’t need 16 GB, you need 32 GB.
Unlikely, at least on non-nvidia chips, and even on AMD, it’s only the latest four chips that support it. Anything older isn’t going to cut it.
You also need a fairly big amount of VRAM for models like that. (4 GB is the minimum for the common kinds, which is more than typical integrated systems, or 8 GB of system memory). You can get by with system RAM, but the performance will be quite bad, since you’re either relying on the CPU, or you’ll be adding the latency from data moving between them.
I haven’t tried LLMs myself, but even completely made up garbage would be better than today’s search engine results.
You either get advertisements for things that have nothing to do with what you’re trying to find or you get privacy preserving links to sites that have nothing to do with what you’re trying to find.
Same reason people get their WiFi from their ISP Modem+Router combo, even though it’s stupid to do so: People often confuse initial convenience for good.
I haven’t used Opera since they switched from their own engine to chrome. They are now owned by a Chinese company, so it probably has at least as much tracking built into it as Google Chrome now.
I used Fifth a bit, which is something aesthetically similar to old Opera made with fltk and a webkit port to fltk. But it’s abandoned now.
It’s so sad really, when I was a Windows user, it was Opera, when I moved to Linux, it was again Opera, then I also started using Conkeror (based on XULRunner).
Then Opera died. Then XULRunner died. No usable web browser anymore.
Have a look at Otter browser It aims to replicate the old interface. It is using QtWebEngine as Presto was closed source. It is in development since 10 years now. And it is open source.
Qt WebEngine uses code from the Chromium project. However, it is not containing all of Chrome/Chromium: Binary files are stripped out Auxiliary services that talk to Google platforms are stripped out, Source
While that’s one of the reasons I don’t want to use chromium, it’s not actually the main reason, if so I’d just use Ungoogled Chromium. I just want more web engines, and I dont want google to monopolise the internet.
It is super hard to create a new web engine, especially when one company is influencing the web standards and most web developers are only testing against that because of market share. This is why we ended up with four active web engines. In alphabetical order: Blink, Gecko, Goanna, WebKit. Obviously some are related: WebKit started out as the fork of KDE’s KHTML and Blink is the fork of WebKit. Goanna is the fork of the Unified XUL Platform that was forked from Mozilla’s Gecko.
Add comment