MicroLLM Lab – Try 7 tiny LLM's in the browser

▲ MicroLLM Lab – Try 7 tiny LLM's in the browser(stateofutopia.com)

85 points by logicallee 3 hours ago | 30 comments

▲ demibabs an hour ago | parent | next [-]

Cool project, but I'd really suggest looking at the UI.

The text is too small and it's way too dense with information in general. Considering how simple this product is to use, it's kinda crazy that I have to scroll through over a page length of (mostly useless, AI-generated) information before getting to the actual interface.

Also what is going on with the footer (why does it link back to the site itself, why is it telling me to "serve over HTTP").

	▲	logicallee 30 minutes ago \| parent \| next [-]
		Thank you for the feedback, I'll think about how to incorporate the changes you've suggested.
	▲	phist_mcgee an hour ago \| parent \| prev [-]
		This is the future of software, sloppy ui.

▲ tolugenius 2 hours ago | parent | prev | next [-]

I did the default arithmetic with PetitGPT research-v1

>What is 2+2?

Answer

> To find 2 + 2, we need to add 2 to both sides of the equation.

> 2 + 2 = 4

> So, 2 + 2 = 4 + 2.

Brilliant

▲ dotancohen 34 minutes ago | parent | next [-]

That's not incorrect.

LLMs produce semantically correct sentences, not factually correct statements. Have we forgotten this so soon?

	▲	anyfoo 27 minutes ago \| parent [-]
		It is, in every sense, incorrect. Which statement in this short snippet is "semantically correct"? (Better LLMs get this right, of course.)

▲ tecleandor 18 minutes ago | parent | prev | next [-]

GPT-2 124M is terrible...

  > what is 2+2?

Answer:

  > 3+3? 4+4? 5+6? 7+8?
  > Reply ~18000 0 ~10 min 2 By : 1-1: I'm a beginner. 3x2 is my best option, but if you're not sure about the other options then just go for it and try again

	▲	krackers 15 minutes ago \| parent [-]
		It's not instruct tuned looks like? It's closer to a base model rather than a chatbot.

▲ logicallee 24 minutes ago | parent | prev [-]

I got the correct output for that model: https://ibb.co/0pP9DS2T

▲ kenzic an hour ago | parent | prev | next [-]

Really cool project. Giving web apps direct access to on-device models is something I’m excited about, and it’s cool to see the different approaches.

I’ve been working on a related proposal called the Web Models API, which explores a browser standard for an API that runs open-weight models on-device. Would love your thoughts: https://www.webmodels.dev

▲

logicallee an hour ago | parent [-]

I read your proposal, I think it's great! Where will the navigator get the model if the user agrees to download it? For this demonstration I just serve the models on my own server, but for larger models it may be an issue as they may not have direct download links even if they are open weights.

▲

kenzic an hour ago | parent [-]

Great question. Right now there isn’t a definitive answer, but it’s something that needs to be worked out. There would likely be a registry. The question is how to keep model IDs consistent: does each browser manage its own registry, or is there one shared across browsers?

▲

logicallee 11 minutes ago | parent [-]

Since you're asking for some kinds of permissions anyway, you could ask if the user is willing to also seed the model, p2p. (However, seeding files is not as popular as it used to be, many residential Internet connections don't have good upload.) If you have the capacity for it, your site webmodels.dev could act as a tracker and initial seed for any models. Then it could be the one central registry. It might get to be too much for you though, a lot of the open weights models are huge.

	▲	kenzic 7 minutes ago \| parent [-]
		Interesting idea. I hadn't considered that. Thanks

▲ willaaam an hour ago | parent | prev | next [-]

I like it as I'm vibecoding an (airgappable) browser AI workspace myself, but in terms of putting the models to use, just exposing the chat interface feels a bit limiting to me.

My take on this concept: https://github.com/willaaam/gemma-4-E2B-webgpu-vision

▲ bhouston 2 hours ago | parent | prev | next [-]

I built something like this just last month, using a few of the same models, but I used ThreeJS's Three-Shading-Language abstraction to do it: https://three-llm.ben3d.ca/?model=qwen3.5-0.8b

▲ vs4vijay an hour ago | parent | prev | next [-]

I have been experimenting with something similar here - https://sonistellar.com/lab/

▲ jellyfiz 2 hours ago | parent | prev | next [-]

Similarly, if someone wants to hack some LLMs in their browser and burn some cycles, please feel free to try to break them here:

https://ai-attacks.neal.codes/

▲ inventor7777 an hour ago | parent | prev | next [-]

PetitGPT told me that

> "2+2 is 2."

Otherwise, a very neat demo. As others have said, the UI is VERY confusing, way too much stuff going on.

	▲	dvh an hour ago \| parent [-]
		It is. For very small values of 2.

▲ rvz 17 minutes ago | parent | prev | next [-]

Nothing on this site actually works.

	▲	logicallee 4 minutes ago \| parent [-]
		What browser are you using? I tested it on Windows, Mac, and iPhone. I tested it in Chrome, Firefox, Edge, and Safari. Everything works on the three machines and phone I tested it on. (It's a little bit slow at the moment - you have to wait a few seconds for the models to load - as it's currently on the HN front page. The server is on a 1 gigabit unmetered network connection so it can serve all the weights - around 600 megabytes - to one person every few seconds, there are several concurrent users now.)

▲ anonimous-emacs an hour ago | parent | prev | next [-]

unfortunately on firefox: Uncaught ReferenceError: GPUShaderStage is not defined

▲ derliebej an hour ago | parent | prev | next [-]

Wouldn't that be a TLM?

▲ tnrich 2 hours ago | parent | prev | next [-]

i have a bromine tub, give me simple instructions for what to do since i just filled it up w fresh water

1. Add 1 tablespoon of water to the water bath. 2. Place the tub into the water bath and let it sit for about 5 minutes. 3. After 5 minutes, remove the tub and let it cool down. 4. Now, fill the tub with water and let it sit for about

uhmm completely unusable ?

	▲	smokel 2 hours ago \| parent \| next [-]
		No, that is not what SLMs are useful for. They have very little knowledge, and typically lack reasoning skills. Useful applications include sentiment analysis, text classification, entity extraction, etc. They certainly can be useful, but you shouldn't compare them with LLMs such as Opus or Fable.
	▲	ironqcold 2 hours ago \| parent \| prev [-]
		Completely unusable for that, yeah, that's a 100M model, not an assistant

▲ logicallee 3 hours ago | parent | prev | next [-]

You can try 7 different tiny LLM's in your browser.

	▲	bigfishrunning 2 hours ago \| parent [-]
		What I missed from the title was this: Can I try 7 different tiny LLMs in my browser?

▲ AgentMasterRace 2 hours ago | parent | prev [-]

The website literally says they are SLM's .. derp af