[HN Gopher] Web LLM
       ___________________________________________________________________
        
       Web LLM
        
       Author : synergy20
       Score  : 75 points
       Date   : 2023-04-25 13:39 UTC (9 hours ago)
        
 (HTM) web link (mlc.ai)
 (TXT) w3m dump (mlc.ai)
        
       | youssefabdelm wrote:
       | Can someone explain why these types of projects are compelling
       | (running an LLM in the user's browser?)
       | 
       | I guess privacy and ease-of-use?
       | 
       | So you don't need to download llama.cpp, and run it in the
       | terminal or something?
        
         | b_mc2 wrote:
         | A reason I like it is I have an "older" AMD GPU which is no
         | longer supported by ROCm (sort of AMDs version of Cuda) which
         | means running locally I'm either trying to figure out older
         | ROCm builds to use my GPU and running into dependency issues or
         | using my CPU which isn't that great either. But with WebGPU I'm
         | able to run these models on my GPU which has been much faster
         | than using the .cpp builds.
         | 
         | Its also fairly easy to route a Flask server to these models
         | with websockets, so with that I've been able to run python and
         | pass data to the model to run on the GPU and pass the response
         | back to the program. Again, there's probably a better way but
         | its cool to have my own personal API for a LLM.
        
         | Veen wrote:
         | Yeah, ease of use. I haven't wanted to bother with running
         | local LLMs until now because I expected it would be complicated
         | and time consuming (and I'm technically competent). I managed
         | to get this up-and-running in about five minutes.
        
           | sourcecodeplz wrote:
           | I thought the same but llama.cpp is literally two downloads
           | in the same folder and then you double click an .exe
        
             | woah wrote:
             | Reminds me of the HN comments on the original dropbox
             | announcement: https://news.ycombinator.com/item?id=8863
        
               | AmericanChopper wrote:
               | I know this comment is a bit of a meme, but in my
               | experience "well actually you can already do this on
               | Linux" is almost universally a calling card of an
               | impending bad opinion.
        
               | tpoacher wrote:
               | Why bad? OP from 2007 still has a point.
               | 
               | Actually, given dropbox's deterioration over time,
               | FTP+SVN is sounding pretty good to me right now.
        
               | AmericanChopper wrote:
               | Because the "do it yourself on linux" advocate tends to
               | ignore all of the reasons why a person might prefer an
               | easy to use, 3rd party managed services. Which could
               | include not wanting to accept all of the compromises
               | required to use Linux for personal computing, not having
               | the competence to implement a Linux solution, or not
               | having the motivation or time to maintain their own
               | Linux-based services. So even when the solutions they
               | advocate are perfectly decent, it tends to be associated
               | with a lack of insight into how small of a segment people
               | with the same competencies and preferences as them
               | represent.
        
             | EGreg wrote:
             | You lost me at .exe
        
         | zhte415 wrote:
         | As well as privacy, cheaper and more scalable for a web service
         | to use a user's resources than server side jumps out at me.
        
       | ccozan wrote:
       | I find theses affirmations a bit contradictory:
       | 
       | "This opens up a lot of fun opportunities to build AI assistants
       | for everyone"
       | 
       | "We have tested it on windows and mac, you will need a gpu with
       | about 6.4G memory."
        
         | capableweb wrote:
         | Yeah, I don't think it's super common to have that available,
         | in most segments.
         | 
         | Although I think Apple doesn't sell computers with less than
         | 8GB RAM, which is used by the GPU as well as the memory is
         | unified.
         | 
         | And latest Valve survey says that at least ~54% of people using
         | Steam who participate in the survey, has a GPU with 8GB or
         | more.
         | 
         | According to https://web3dsurvey.com/webgpu WebGPU enjoys a
         | whopping ~6% browser support.
         | 
         | But the previous two segments are very small ones, if you
         | compare to everywhere computers are being used. Coupled
         | together with that WebGPU is hardly enabled anywhere, I
         | wouldn't be surprised if maybe only ~1% of everyone using their
         | own computer have access to 8GB VRAM + can use WebGPU.
         | 
         | So yeah, bit weird to say everyone when looking at the
         | requirements of 8GB VRAM + WebGPU enabled browser.
        
           | flohofwoe wrote:
           | > WebGPU enjoys a whopping ~6% browser support.
           | 
           | That's actually not too bad considering that WebGPU hasn't
           | been released yet (the number will drastically go up when
           | Chrome 113 is out of beta)
        
             | capableweb wrote:
             | > the number will drastically go up when Chrome 113 is out
             | of beta
             | 
             | Which coincidentally is happening in about 5 hours (but
             | it'll take many days for the update to actually go
             | everywhere) according to https://chromestatus.com/roadmap,
             | exciting :)
        
       ___________________________________________________________________
       (page generated 2023-04-25 23:02 UTC)