_______ __ _______
| | |.---.-..----.| |--..-----..----. | | |.-----..--.--.--..-----.
| || _ || __|| < | -__|| _| | || -__|| | | ||__ --|
|___|___||___._||____||__|__||_____||__| |__|____||_____||________||_____|
on Gopher (inofficial)
URI Visit Hacker News on the Web
COMMENT PAGE FOR:
DIR Ask HN: Is anyone doing anything cool with tiny language models?
lightning19 wrote 22 hours 36 min ago:
not sure if it is cool but, purely out of spite, I'm building a LLM
summarizer app to compete with a AI startup that I interviewed with.
The founders were super egotistical and initially thought I was not
worthy of an interview.
herol3oy wrote 23 hours 3 min ago:
I've created Austen [0] to generate relationships between book
characters using Mermaid.
[0]
URI [1]: https://github.com/herol3oy/austen
sebazzz wrote 23 hours 17 min ago:
I built auto-summarization and grouping in an experimental branch of my
hobby-retrospective tool: [1] Iâm now just wondering if there is any
way to build tests on the input+output of the LLM :D
URI [1]: https://github.com/Sebazzz/Return/tree/experiment/ai-integrati...
panchicore3 wrote 23 hours 51 min ago:
I am moderating a playlists manager to restrict them to a range of
genders so it classifies song requests as accepted/rejected.
accrual wrote 1 day ago:
Although there are better ways to test, I used a 3B model to speed up
replies from my local AI server when testing out an application I was
developing. Yes I could have mocked up HTTP replies etc., but in this
case the small model let me just plug in and go.
Thews wrote 1 day ago:
Before ollama and the others could do structured JSON output, I hacked
together my own loop to correct the output. I used it that for dummy
API endpoints to pretend to be online services but available locally,
to pair with UI mockups. For my first test I made a recipe generator
and then tried to see what it would take to "jailbreak" it. I also
used uncensored models to allow it to generate all kinds of funny
content.
I think the content you can get from the SLMs for fake data is a lot
more engaging than say the ruby ffaker library.
bashbjorn wrote 1 day ago:
I'm working on a plugin[1] that runs local LLMs from the Godot game
engine. The optimal model sizes seem to be 2B-7B ish, since those will
run fast enough on most computers. We recommend that people try it out
with Gemma 2 2B (but it will work with any model that works with
llama.cpp)
At those sizes, it's great for generating non-repetitive flavortext for
NPCs. No more "I took an arrow to the knee".
Models at around the 2B size aren't really capable enough to act a
competent adversary - but they are great for something like bargaining
with a shopkeeper, or some other role where natural language can let
players do a bit more immersive roleplay.
URI [1]: https://github.com/nobodywho-ooo/nobodywho
Tepix wrote 23 hours 15 min ago:
Cool. Are you aware of good games that use LLMs like this?
aDyslecticCrow wrote 21 hours 25 min ago:
I have not seen much myself, but it's one of the earliest use cases
I thought about when they started showing up.
jbentley1 wrote 1 day ago:
Tiny language models can do a lot if they are fine tuned for a specific
task, but IMO a few things are holding them back:
1. Getting the speed gains is hard unless you are able to pay for
dedicated GPUs. Some services offer LoRA as serverless but you don't
get the same performance for various technical reasons.
2. Lack of talent to actually do the finetuning. Regular engineers can
do a lot of LLM implementation, but when it comes to actually
performing training it is a scarcer skillset. Most small to medium orgs
don't have people who can do it well.
3. Distribution. Sharing finetunes is hard. HuggingFace exists, but
discoverability is an issue. It is flooded with random models with no
documentation and it isn't easy to find a good oen for your task. Plus,
with a good finetune you also need the prompt and possibly parsing code
to make it work the way it is intended and the bundling hasn't been
worked out well.
grisaitis wrote 1 day ago:
when you say fine-tuning skills or talent are scarce, do you have
specific skills in mind? perhaps engineering for training models (eg
making model parallelism work)? or the more ML type skills of
designing experiments, choosing which methods to use, figuring out
datasets for training, hyperparam tuning/evaluation, etc?
jbentley1 wrote 1 day ago:
The technical parts are less common and specialized, like
understanding the hyperparameters and all that, but I don't think
that is the main problem. Most people don't understand how to build
a good dataset or how to evaluate their finetune after training.
Some parts of this are solid rules like always use a separate
validation set, but the task dependent parts are harder to teach.
It's a different problem every time.
menaerus wrote 6 hours 0 min ago:
Finetuning, as I understand it, is mostly laborious and mostly
very boring and exhausting work that is not appealing to many
engineers. It can be done by people who have some skills in
Python or similar language and who have some background in
statistics.
OTOH to build the infra for LLMs there's much more stuff involved
and it's really hard to find engineers who have the capacity to
be both the researchers and developers at the same time. By
"researchers" I mean that they have to have a capacity to be able
to read through the numerous academic and industry papers,
comprehend the tiniest details, and materialize it into the
product through the code. I think that's much harder and scarcer
skill to find.
That said, I am not undermining the fine-tuning skill, it's a
humongous effort, but I think it's not necessarily the skillset
problem.
mogaal wrote 1 day ago:
I bought a tiny business in Brazil, the database (Excel) I inherited
with previous customer data *do not include gender*. I need gender to
start my marketing campaigns and learn more about my future customer. I
used Gemma-2B and Python to determine gender based on the data and it
worked perfect
Nashooo wrote 1 day ago:
How did you verify it worked?
addandsubtract wrote 1 day ago:
I use a small model to rename my Linux ISOs. I gave it a custom prompt
with examples of how I want the output filenames to be structured and
then just feed it files to rename. The output only works 90ish percent
of the time, so I wrote a little CLI to iterate through the files and
accept / retry / edit the changes the LLM outputs.
ahrjay wrote 1 day ago:
I built [1] using the chrome ai (Gemini nano). Allows you to do ffmpeg
operations on videos using natural language all client side.
URI [1]: https://ffprompt.ryanseddon.com
fauigerzigerk wrote 1 day ago:
What are the prerequisites for this? I keep getting "Bummer, looks
like your device doesn't support Chrome AI" on macOS 15.2 Chrome
132.0.6834.84 (Official Build) (arm64)
[Edit] Found it. I had to enable
chrome://flags/#prompt-api-for-gemini-nano
krystofee wrote 1 day ago:
Has anyone ever tried to do some automatic email workflow autoresponder
agents?
Lets say, I want some outcome and it will autonomousl handle the
process prompt me and the other side for additional requirements if
necessary and then based on that handle the process and reach the
outcome?
lormayna wrote 1 day ago:
I am using smollm2 to extract some useful information (like remote,
language, role, location, etc.) from "Who is hiring" monthly thread and
create an RSS feed with specific filter.
Still not ready for Show HN, but working.
reeeeee wrote 1 day ago:
I built a platform to monitor LLMs that are given complete freedom in
the form of a Docker container bash REPL. Currently the models have
been offline for some time because I'm upgrading from a single DELL to
a TinyMiniMicro Proxmox cluster to run multiple small LLMs locally.
The bots don't do a lot of interesting stuff though, I plan to add the
following functionalities:
- Instead of just resetting every 100 messages, I'm going to provide
them with a rolling window of context.
- Instead of only allowing BASH commands, they will be able to also
respond with reasoning messages, hopefully to make them a bit smarter.
- Give them a better docker container with more CLI tools such as curl
and a working package manager.
If you're interested in seeing the developments, you can subscribe on
the platform!
URI [1]: https://lama.garden
tomholandpick wrote 1 day ago:
How accurate? are the classifications?
kolinko wrote 1 day ago:
Appleâs on device models are
around 3B if Iâm nit mistaken, and they developed some nice tech
around them that they published, if Iâm not
mistaken - where they have just one model, but have switchable
finetunings of that model so that it can perform different
functionalities depending on context.
numba888 wrote 1 day ago:
Many interesting projects, cool. I'm waiting to LLMs in games. That
would make them much more fun. Any time now...
jaggs wrote 20 hours 30 min ago:
Have you seen a AI People?
URI [1]: https://www.aipeoplegame.com/
ceritium wrote 1 day ago:
I am doing nothing, but I was wondering if it would make sense to
combine a small LLM and SQLITE to parse date time human expressions.
For example, given a human input like "last day of this month", the LLM
will generate the following query `SELECT date('now','start of
month','+1 month','-1 day');`
It is probably super overengineering, considering that pretty good
libraries are already doing that on different languages, but it would
be funny. I did some tests with chatGPT, and it worked sometimes. It
would probably work with some fine-tuning, but I don't have the
experience or the time right now.
TachyonicBytes wrote 1 day ago:
What libraries have you seen that do this?
lionkor wrote 1 day ago:
LLMs tend to REALLY get this wrong. Ask it to generate a query to sum
up likes on items uploaded in the last week, defined as the last
monday-sunday week (not the last 7 days), and watch it get it subtly
wrong almost every time.
computers3333 wrote 1 day ago:
[1] â I built GopherSignal!
It's a lightweight tool that summarizes Hacker News articles. For
example, hereâs what it outputs for this very post, "Ask HN: Is
anyone doing anything cool with tiny language models?":
"A user inquires about the use of tiny language models for interesting
applications, such as spam filtering and cookie notice detection. A
developer shares their experience with using Ollama to respond to SMS
spam with unique personas, like a millennial gymbro or a 19th-century
British gentleman. Another user highlights the effectiveness of 3B and
7B language models for cookie notice detection, with decent performance
achieved through prompt engineering."
I originally used LLaMA 3:Instruct for the backend, which performs much
better, but recently started experimenting with the smaller LLaMA
3.2:1B model.
Itâs been cool seeing other peopleâs ideas too. Curiousâdoes
anyone have suggestions for small models that are good for summaries?
Feel free to check it out or make changes:
URI [1]: https://gophersignal.com
URI [2]: https://github.com/k-zehnder/gophersignal
tinco wrote 1 day ago:
That's cool, I really like it. One piece of feedback: I am usually
more interested in the HN comments than in the original article. If
you'd include a link to the comments then I might switch to
GopherSignal as a replacement for the HN frontpage.
My flow is generally: Look at the title and the amount of upvotes to
decide if I'm interested in the article. Then view the comments to
see if there's interesting discussion going on or if there's already
someone adding essential context. Only then I'll decide if I want to
read the article or not.
Of course no big deal if you're not interested in my patronage, just
wanted to let you know your page already looks good enough for me to
consider switching my most visited page to it if it weren't for this
small detail. And maybe the upvote count.
computers3333 wrote 8 hours 3 min ago:
EDIT: Apologies for breaking things earlier while trying to fix it!
Iâve been working on updating it and got the upvote count and
comment link in there. Wondering what you think about these
updatesâappreciate any feedback! Thanks again for helping me
improve it!
URI [1]: https://gophersignal.com
goodklopp wrote 22 hours 8 min ago:
I would love this feature. Regardless, what you have built is
really cool
computers3333 wrote 21 hours 57 min ago:
Hey thanks a ton for checking out GopherSignal! From the feedback
Iâm getting, it seems like comments and upvotes are the secret
sauce Iâve been missingâappreciate you helping me get that
through my thick skull lol. The pressureâs on nowâIâll do
my best to deliver.
sainib wrote 1 day ago:
May be even rate each post on the comments activity level.
computers3333 wrote 22 hours 3 min ago:
Great call! Thatâs a really solid ideaâusing the LLMs to rate
posts based on comment activity could totally work and would be
fun.
Were you thinking something like a âDramaLlama,â deciding if
itâs a slow day or a meltdown-worthy soap opera in the
comments? Or maybe something more valuable, like an âInsight
Indexâ that uses the LLM to analyze comments for links,
explanations, or phrases that add context or insightâbasically
gauging how constructive or meaningful the discussion is?
I also saw an idea in another post on this thread about an LLM
that constantly listens to conversations and declares a winner.
That could be fun to adapt for spicier postsâlike the LLM
picking a âwinnerâ in the comments. Make the argument
GopherSignal official lol. If it helps bring in another user,
Iâm all in!
Appreciate the feedback.
sainib wrote 1 day ago:
Agreed..great suggestions. Id consider switching as well.
computers3333 wrote 1 day ago:
Hey, thanks a ton for the feedback! That was super helpful to hear
about your flowâmakes a lot of sense and it's pretty similar to
how I browse HN too. I usually only dive into the article after
checking out the upvotes and seeing what context the comments add.
I'll definitely add a link to the comments and the upvote
countâgotta keep my tiny but mighty userbase (my mom, me, and
hopefully you soon) happy, right? lol
And if there's even a chance you'd use GopherSignal as your daily
driver, that's a no-brainer for me. Really appreciate you taking
the time to share your ideas and help me improve.
evacchi wrote 1 day ago:
I'm interested in finding tiny models to create workflows stringing
together several function/tools and running them on device using
mcp.run servlets on Android (disclaimer: I work on that)
kaspermarstal wrote 1 day ago:
I built an Excel Add-In that allows my girlfriend to quickly filter
7000 paper titles and abstracts for a review paper that she is writing
[1]. It uses Gemma 2 2b which is a wonderful little model that can run
on her laptop CPU. It works surprisingly well for this kind of binary
classification task.
The nice thing is that she can copy/paste the titles and abstracts in
to two columns and write e.g. "=PROMPT(A1:B1, "If the paper studies
diabetic neuropathy and stroke, return 'Include', otherwise return
'Exclude'")" and then drag down the formula across 7000 rows to bulk
process the data on her own because it's just Excel. There is a gif on
the readme on the Github repo that shows it.
URI [1]: https://github.com/getcellm/cellm
basmok wrote 20 hours 35 min ago:
Can someone hack this together as pure matrix multiplication?
Like either as table in the background or as regular script?
On most computers you can't compile or add add-ons without
administrative rights and LLM Chat sites are blocked to prevent usage
of company data.
It should run on native Excel or GSheets.
I mean, pure without compilation, just like the do the matrix
calculations here straight in Excel without admin rights:
Lesson 1: Demystifying how LLMs work, from architecture to Excel [1]
As far as i know in GSheet the scripts also run on the Google Servers
and are not limited by the local computer power. So there larger
models could be deployed.
Someone can hack this into Excel/GSheet?
URI [1]: https://youtu.be/FyeN5tXMnJ8
vdm wrote 1 day ago:
URI [1]: https://x.com/Suhail/status/1882069209129340963
7734128 wrote 1 day ago:
You could have called it CellMate
7734128 wrote 1 day ago:
You could have called it CellMate b
donbreo wrote 1 day ago:
Requirements:
-Windows
Looks like I'm out...
Would be great if there was a google apps script alternative. My
company gave all devs linux systems and the business team operates on
windows. So I always use browser based tech like Gapps script for
complex sheet manipulation
jkman wrote 22 hours 54 min ago:
Well it's an excel add-in, how else would it work?
NotMichaelBay wrote 3 hours 14 min ago:
Excel add-ins can be written with the Office JS API so that they
can run on web as well as desktop for Windows and Mac. But I
don't think OP's add-in is possible with that API unless the
local model can be run in JS.
afro88 wrote 1 day ago:
How accurate are the classifications?
kaspermarstal wrote 1 day ago:
I don't know. This paper [1] reports accuracies in the 97-98% range
on a similar task with more powerful models. With Gemma 2 2b the
accuracy will certainly be lower.
URI [1]: https://www.medrxiv.org/content/10.1101/2024.10.01.2431470...
beernet wrote 1 day ago:
> I don't know.
HN in a nutshell: I've built some cool tech but have no idea if
it is helpful or even counter productive...
kaspermarstal wrote 22 hours 55 min ago:
I am not going to claim or report any kind of accuracy,
especially with such a small model and such a specific, context
dependent use case. It is the userâs responsibility to cross
validate if itâs accurate enough for their use case and
upgrade model or use another approach if not.
jbs789 wrote 22 hours 19 min ago:
A user buys a car because it gets them from point A to point
B. I get what youâre saying though - we are earlier along
the adoption curve for these models and more responsibility
sits with the user. Over time the expectations will no doubt
increase.
sidcool wrote 23 hours 11 min ago:
Sometimes it's the joy of creation. Utility and optimization
come later. It's fun. Like a hobby.
corobo wrote 1 day ago:
Real HN in a nutshell: People who don't build stuff telling
people who do build stuff that the thing they built is useless
:P
It's a hacker forum, let people hack!
If anything have a dig at OP for posting the thread too soon
before the parent commenter has had the chance to gather any
data, haha
greenavocado wrote 1 day ago:
Just because you can, doesn't mean you should
corobo wrote 23 hours 56 min ago:
If you're building a dinosaur sanctuary sure
stackghost wrote 22 hours 6 min ago:
Or an Internet surveillance-capitalism panopticon.
rasmus1610 wrote 1 day ago:
Sometimes people just like to build stuff for the sake of it.
jajko wrote 1 day ago:
Almost like hackers, doing shit just for the heck of it
because they can (mostly)
indolering wrote 1 day ago:
Y'all definitely need to cross validate a small number of samples
by hand. When I did this kind of research, I would hand validate
to at least P < .01.
kaspermarstal wrote 1 day ago:
She and one other researcher has manually classified all 7000
papers as per standard protocol. Perhaps for the next article
they will measure how this tool agreed with them against them
and include it in the protocol if good enough.
relistan wrote 1 day ago:
Very cool idea. Iâve used gemma2 2b for a few small things. Very
good model for being so small.
merwijas wrote 1 day ago:
I put llama 3 on a RBPi 5 and have it running a small droid. I added a
TTS engine so it can hear spoken prompts which it replies to in droid
speak. It also has a small screen that translates the response to
English. I gave it a backstory about being a astromech droid so it
usually just talks about the hyperdrive but it's fun.
sauravpanda wrote 1 day ago:
We are building a framework to run this tiny language model in the web
so anyone can access private LLMs in their browser: [1] .
With just three lines of code, you can run Small LLM models inside the
browser. We feel this unlocks a ton of potential for businesses so that
they can introduce AI without fear of cost and can personalize the
experience using AI.
Would love your thoughts and what we can do more or better!
URI [1]: https://github.com/sauravpanda/BrowserAI
ms7892 wrote 1 day ago:
Sounds cool. Anyway I can help.
guywithahat wrote 1 day ago:
I've been working on a self-hosted, low-latency service for small
LLM's. It's basically exactly what I would have wanted when I started
my previous startup. The goal is for real time applications, where even
the network time to access a fast LLM like groq is an issue.
I haven't benchmarked it yet but I'd be happy to hear opinions on it.
It's written in C++ (specifically not python), and is designed to be a
self-contained microservice based around llama.cpp.
URI [1]: https://github.com/thansen0/fastllm.cpp
gpm wrote 1 day ago:
I made a shell alias to translate things from French to English, does
that count?
function trans
llm "Translate \"$argv\" from French to English please"
end
Llama 3.2:3b is a fine French-English dictionary IMHO.
kreyenborgi wrote 1 day ago:
Is it better than translatelocally? [1] (the same as used in firefox)
URI [1]: https://translatelocally.com/downloads/
gpm wrote 1 day ago:
It's different. It doesn't always just give one translation but
different options. I can do things like give it a phrase and then
ask it to break it down. Or give it a word and if its translation
doesn't make sense to me ask how it works in the context of a
phrase.
llm -c, which continues the previous conversation, is specifically
useful for that sort of manipulation.
It's also available from the command line, which I find convenient
because I basically always have one open.
dh1011 wrote 1 day ago:
I copied all the text from this post and used an LLM to generate a list
of all the ideas. I do the same for other similar HN post .
whalesalad wrote 1 day ago:
chatgpt did a stellar job parsing the "books on hard things" thread
from a little while ago. my prompt was:
Can you identify all the books here, sorted by a weight which is
determined based on a combo of the number of votes the comment has,
the number of sub-comments, or the number of repeat mentions.
Ideally retain hyperlinks if possible.
swifthesitation wrote 8 hours 41 min ago:
could you link the HN thread?
whalesalad wrote 1 hour 43 min ago:
google "hn books on hard things" -
URI [1]: https://news.ycombinator.com/item?id=42614722
lordswork wrote 1 day ago:
well, what are the ideas?
sidravi1 wrote 1 day ago:
We fine-tuned a Gemma 2B to identify urgent messages sent by new and
expecting mothers on a government-run maternal health helpline.
URI [1]: https://idinsight.github.io/tech-blog/blog/enhancing_maternal_...
Mukina wrote 9 hours 6 min ago:
Super cool. What a simple and powerful way to help mothers in need.
Thanks for sharing.
Mumps wrote 1 day ago:
lovely application!
Genuine question: why not use (Modern)BERT instead for
classification? (Is the json-output explanation so critical?)
Mashimo wrote 1 day ago:
Oh that is a nice writeup. We have something similar in mind at work.
Will forward it.
proxygeek wrote 1 day ago:
Such a fun thread but this is the kind of applications that perk up
my attention!
Very cool!
linsomniac wrote 1 day ago:
I have this idea that a tiny LM would be good at canonicalizing entered
real estate addresses. We currently buy a data set and software from
Experian, but it feels like something an LM might be very good at.
There are lots of weirdnesses in address entry that regexes have a hard
time with. We know the bulk of addresses a user might be entering,
unless it's a totally new property, so we should be able to train it on
that.
thesz wrote 16 hours 11 min ago:
From my experience (2018), run LLM output through beam search over
different choices of canonicalization of certain part of text. Even
3-gram models (yeah, 2018) fare better this way.
jftuga wrote 1 day ago:
I'm using ollama, llama3.2 3b, and python to shorten news article
titles to 10 words or less. I have a 3 column web site with a list of
news articles in the middle column. Some of the titles are too long
for this format, but the shorter titles appear OK.
HexDecOctBin wrote 1 day ago:
Is there any experiments in a small models that does paraphrasing? I
tried hsing some off-the-shelf models, but it didn't go well.
I was thinking of hooking them in RPGs with text-based dialogue, so
that a character will say something slightly different every time you
speak to them.
krystofee wrote 1 day ago:
Intuitively this sounds like something that should be possible using
almost any llm. This should be just a matter of prompting.
jwitthuhn wrote 1 day ago:
I've made a tiny ~1m parameter model that can generate random Magic the
Gathering cards that is largely based on Karpathy's nanogpt with a few
more features added on top.
I don't have a pre-trained model to share but you can make one yourself
from the git repo, assuming you have an apple silicon mac.
URI [1]: https://github.com/jlwitthuhn/TCGGPT
itskarad wrote 1 day ago:
I'm using ollama for parsing and categorizing scraped jobs for a local
job board dashboard I check everyday.
JLCarveth wrote 1 day ago:
I used a small (3b, I think) model plus tesseract.js to perform OCR on
an image of a nutritional facts table and output structured JSON.
ian_zcy wrote 1 day ago:
what are you feed into the model? Image (like product packaging) or
Image of Structured Table? I found out that model performs good in
general with sturctured table, but fails a lot over images.
tigrank wrote 1 day ago:
All that server side or client?
deivid wrote 1 day ago:
What was the model? What kind of performance did you get out of it?
Could you share a link to your project, if it is public?
JLCarveth wrote 1 day ago:
[1] I've had good speed / reliability with TheBloke/rocket-3B-GGUF
on Huggingface, the Q2_K model. I'm sure there are better models
out there now, though.
It takes ~8-10 seconds to process an image on my M2 Macbook, so not
quite quick enough to run on phones yet, but the accuracy of the
output has been quite good.
URI [1]: https://github.com/JLCarveth/nutrition-llama
codazoda wrote 1 day ago:
I had an LLM create a playlist for me.
Iâm tired of the bad playlists I get from algorithms, so I made a
specific playlist with an Llama2 based on several songs I like. I
started with 50, removed any I didnât like, and added more to fill in
the spaces. The small models were pretty good at this. Now I have a
decent fixed playlist. It does get âtiredâ after a few weeks and I
need to add more to it. Iâve never been able to do this myself with
more than a dozen songs.
DonHopkins wrote 23 hours 49 min ago:
How about having an LLM create a praylist for you?
Then you could implement Salvation as a Service, where you privately
confess your sins to a local LLM, and it continuously prays for your
eternal soul, recommends penances, and even recites Hail Marys for
you.
jamesponddotco wrote 1 day ago:
Interesting! I wrote a prompt for something similar[1], but I use
Claude Sonnet for it. I wonder how a small model would handle it.
Time to test, I guess.
[1]
URI [1]: https://git.sr.ht/~jamesponddotco/llm-prompts/tree/trunk/dat...
codazoda wrote 19 hours 9 min ago:
This prompt is a lot more complex than what I did. I donât
recall my exact prompt but it was something like, âGenerate a
list of 25 songs that I may like if I like Girl is on my Mind by
the Black Keys.â
Mashimo wrote 1 day ago:
Huh, interesting. For me that often dreamed up artist and songs.
petesergeant wrote 1 day ago:
Interesting! I've sadly found more capable models to really fail on
music recommendations for me.
kianN wrote 1 day ago:
I donât know if this counts as tiny but I use llama 3B in prod for
summarization (kinda).
Its effective context window is pretty small but I have a much more
robust statistical model that handles thematic extraction. The llm is
essentially just rewriting ~5-10 sentences into a single paragraph.
Iâve found the less you need the language model to actually do, the
less the size/quality of the model actually matters.
jothflee wrote 1 day ago:
when i feel like casually listening to something, instead of
netflix/hulu/whatever, i'll run a ~3b model (qwen 2.5 or llama 3.2) and
generate and audio stream of water cooler office gossip. (when it is
up, it runs here: [1] ).
some of the situations get pretty wild, for the office :)
URI [1]: https://water-cooler.jothflee.com
jaggs wrote 20 hours 37 min ago:
I love it. It's a shame the voices aren't just a little bit more
realistic. There's some good models and tts around now I wonder if
you could upgrade it?
jftuga wrote 1 day ago:
What prompt are you using for this?
spiritplumber wrote 1 day ago:
My husband and me made a stock market analysis thing that gets it right
about 55% of the time, so better than a coin toss. The problem is that
it keeps making unethical suggestions, so we're not using it to trade
stock. Does anyone have any idea what we can do with that?
febed wrote 1 day ago:
What data do you analyze?
bongodongobob wrote 1 day ago:
You can literally flip coins and get better than 50% success in a
bull market. Just buy index funds and spend your time on something
that isn't trying to beat entropy. You won't be able to.
spiritplumber wrote 1 day ago:
INSUFFICIENT DATA FOR A MEANINGFUL ANSWER.
dkga wrote 1 day ago:
Suggestion: calculate the out-of-sample Sharpe ratio[0] of the
suggestions over a reasonable period to gauge how good the model
would actually perform in terms of return compared to risks. It is
better than vanilla accuracy or related metrics. Source: I'm a
financial economist.
[0]:
URI [1]: https://en.wikipedia.org/wiki/Sharpe_ratio
spiritplumber wrote 1 day ago:
thank you! that's exactly the sort of thing I don't know.
Etheryte wrote 1 day ago:
Have you backtested this in times when markets were not constantly
green? Nearly any strategy is good in the good times.
spiritplumber wrote 1 day ago:
yep. the 55% is over a few years.
kortilla wrote 1 day ago:
Right, but if 55% is avg over the last few years, âbuy stockâ
is going to be correct more than not.
URI [1]: https://www.crestmontresearch.com/docs/Stock-Yo-Yo.pdf
Etheryte wrote 1 day ago:
I think this is a good highlight of why context and reality
checks are incredibly important when doing work like this. At
first glance, it might look like 55% is a really good result,
but in the previous year, a flat buy every day strategy
would've been right 56.7% of the time.
dutchbookmaker wrote 17 hours 22 min ago:
55% means basically nothing in this context if even money.
Long 45% to 55% is most likely completely random because it
is symmetric with shorting 45% to 55%
Exactly what you would expect from a language model making
random stock picks.
bobbygoodlatte wrote 1 day ago:
I'm curious what sort of unethical suggestions it's coming up with
haha
spiritplumber wrote 1 day ago:
so far, mostly buying companies owned/ran by horrible people.
GordonS wrote 1 day ago:
Can't you adjust the prompt to filter out companies that fund
genocide etc?
kortilla wrote 1 day ago:
So if you filter out the Republican owned ones or whatever your
bugbear is, does the 55% persist?
danbmil99 wrote 1 day ago:
Using llama 3.2 as an interface to a robot. If you can get the latency
down, it works wonderfully
mentos wrote 1 day ago:
Would love to see this applied to a FPS bot in unreal engine.
jmward01 wrote 1 day ago:
I think I am. At least I think I'm building things that will enable
much smaller models:
URI [1]: https://github.com/jmward01/lmplay/wiki/Sacrificial-Training
juancroldan wrote 1 day ago:
I'm making an agent that takes decompiled code and tries to understand
the methods and replace variables and function names one at a time.
krystofee wrote 1 day ago:
This sounds cool! Are you planningto opensource it?
DonHopkins wrote 23 hours 53 min ago:
No need to: he can just publish a binary then you can run it on
itself. ;)
Evidlo wrote 1 day ago:
I have ollama responding to SMS spam texts. I told it to feign
interest in whatever the spammer is selling/buying. Each number gets
its own persona, like a millennial gymbro or 19th century British
gentleman. [1]
URI [1]: http://files.widloski.com/image10%20(1).png
URI [2]: http://files.widloski.com/image11.png
potatoman22 wrote 21 hours 18 min ago:
You probably just get more spam texts since you're replying. Maybe
that's a good thing tbh
bripkens wrote 21 hours 45 min ago:
You should put all these interactions on the web. For education
purposes ofc.
lacoolj wrote 21 hours 52 min ago:
Most spam are just verifying you exist as a person, then from there
you become an actual "target" if you respond.
This feels like an in-between that both wastes their time and adds
you to extra lists.
Send the results somewhere! Not sure if "law enforcement" is
applicable (as in, would be able/willing to act on the info) but if
so, that's a great use of this data :)
hackergirl88 wrote 23 hours 17 min ago:
Where was this during the election
merpkz wrote 1 day ago:
Calling Jessica an old chap is quite a giveaway that it's a bot xD
Nice idea indeed, but I have a feeling that it's just two LLMs now
conversing with each other.
metadat wrote 1 day ago:
I love this, more please!!!
blackeyeblitzar wrote 1 day ago:
You realize this is going to cause carriers to allow the number to
send more spam, because it looks like engagement. The best thing to
do is to report the offending message to 7726 (SPAM) so the carrier
can take action. You can also file complaints at the FTC and FCC
websites, but that takes a bit more effort.
thegabriele wrote 1 day ago:
Yes, the very last thing to do is respond to spam (calls, email,
text...) and inform that you are eligible to more solicitation.
thecosmicfrog wrote 1 day ago:
Please tell me you have a blog/archive of these somewhere. This was
such a joy to read!
celestialcheese wrote 1 day ago:
Given the source, I'm skeptical it's not just a troll, but found this
explanation [0] plausible as to why those vague spam text exists. If
true, this trolling helps the spammers warm those phone numbers up.
0 -
URI [1]: https://x.com/nikitabier/status/1867029883387580571
stogot wrote 1 day ago:
Why does STOP work here?
yawgmoth wrote 1 day ago:
STOP works thanks to the Telephone Consumer Protection Act
(âTCPAâ), which offers consumers spam protections and senders
a framework on how to behave.
(Edit: It's relevant that STOP didn't come from the TCPA itself,
but definitely has teeth due to it)
URI [1]: https://www.infobip.com/blog/a-guide-to-global-sms-compl...
inerte wrote 1 day ago:
Carriers and SMS service providers (like Twillio) obey that, no
matter what service is behind.
There are stories of people replying STOP to spam, then never
getting a legit SMS because the number was re-used by another
service. That's because it's being blocked between the spammer
and the phone.
celestialcheese wrote 1 day ago:
[1] Again, no clue if this is true, but it seems plausible.
URI [1]: https://x.com/nikitabier/status/1867069169256308766
zx8080 wrote 1 day ago:
Cool! Do you consider the risk of unintentional (and until some
moment, an unknown) subscription to some paid SMS service and how do
you mitigate it?
Evidlo wrote 1 day ago:
I have to whitelist a conversation before the LLM can respond.
RVuRnvbM2e wrote 1 day ago:
This is fantastic. How have your hooked up a mobile number to the
llm?
Evidlo wrote 1 day ago:
Android app that forwards to a Python service on remote workstation
over MQTT. I can make a Show HN if people are interested.
SuperHeavy256 wrote 1 day ago:
I am so SO interested, please make a Show HN
Evidlo wrote 20 hours 31 min ago:
URI [1]: https://news.ycombinator.com/item?id=42796496
gaudystead wrote 20 hours 18 min ago:
Sweeeeet, thank you! :)
sainib wrote 1 day ago:
Interested for sure.
potamic wrote 1 day ago:
Why MQTT over HTTP for a low volume, small scale integration?
c0wb0yc0d3r wrote 1 day ago:
Iâm not OP, but I would hazard a guess that those are the
tools that OP has at hand.
dkga wrote 1 day ago:
Yes, I'd be interested in that!
deadbabe wrote 1 day ago:
Iâd love to see that. Could you simulate iMessage?
great_psy wrote 1 day ago:
Yes itâs possible, but itâs not something you can easily
scale.
I had a similar project a few years back that used OSX
automations and Shortcuts and Python to send a message everyday
to a friend. It required you to be signed in to iMessage on
your MacBook.
Than was a send operation, the reading of replies is not
something I implemented, but I know there is a file somewhere
that holds a history of your recent iMessages. So you would
have to parse it on file update and that should give you the
read operation so you can have a conversation.
Very doable in a few hours unless something dramatic changed
with how the messages apps works within the last few years.
dewey wrote 1 day ago:
They are all in a SQLite db on your disk.
Evidlo wrote 1 day ago:
If you mean hook this into iMessage, I don't know. I'm willing
to bet it's way harder though because Apple
dambi0 wrote 1 day ago:
If you are willing to use Apple Shortcuts on iOS itâs
pretty easy to add something that will be trigged when a
message is received and can call out to a service or even use
SSH to do something with the contents, including replying
spiritplumber wrote 1 day ago:
For something similar with FB chat, I use Selenium and run it on
the same box that the llm is running on. Using multiple
personalities is really cool though. I should update mine likewise!
antonok wrote 1 day ago:
I've been using Llama models to identify cookie notices on websites,
for the purpose of adding filter rules to block them in EasyList
Cookie. Otherwise, this is normally done by, essentially, manual
volunteer reporting.
Most cookie notices turn out to be pretty similar, HTML/CSS-wise, and
then you can grab their `innerText` and filter out false positives with
a small LLM. I've found the 3B models have decent performance on this
task, given enough prompt engineering. They do fall apart slightly
around edge cases like less common languages or combined cookie notice
+ age restriction banners. 7B has a negligible false-positive rate
without much extra cost. Either way these things are really fast and
it's amazing to see reports streaming in during a crawl with no human
effort required.
Code is at [1] . You can see the prompt at [1]
/blob/main/src/text-cl....
URI [1]: https://github.com/brave/cookiemonster
URI [2]: https://github.com/brave/cookiemonster/blob/main/src/text-clas...
rpastuszak wrote 1 day ago:
Tangentially related, I worked on something similar, using LLMs to
find and skip sponsored content in YT videos:
URI [1]: https://butter.sonnet.io/
GardenLetter27 wrote 1 day ago:
It's funny that this is even necessary though - that great EU
innovation at work.
vvillena wrote 20 hours 49 min ago:
Bear in mind, those arcane cookie forms are probably not compliant
with EU laws. If there's not a "reject" button next to the "accept"
button, the form is almost definitely not to spec.
pornel wrote 22 hours 39 min ago:
The legislation has been watered down by lobbying of the
trillion-dollar tracking industry.
The industry knows ~nobody wants to be tracked, so they don't want
to let tracking preferences to be easy to express. They want cookie
notices to be annoying to make people associate privacy with a
bureaucratic nonsense, and stop demanding to have privacy.
There was P3P spec in 2002: [1] It even got decent implementation
in Internet Explorer, but Google has been deliberately sending a
junk P3P header to bypass it.
It has been tried again with a very simple DNT spec. Support for it
(that barely existed anyway) collapsed after Microsoft decided to
make Do-Not-Track on by default in Edge.
URI [1]: https://www.w3.org/TR/P3P/
kalaksi wrote 1 day ago:
Tracking, tracking cookies, banners etc. are a choice done by the
website. There are browser addons for making it simpler, though.
The transparency requirements and consent for collecting all kinds
of PII (this is the regulation) actually is a great innovation.
docmars wrote 23 hours 37 min ago:
I think I'd rather see cookie notices handled by a browser API
with a common UI, where the default is always "No." Provide that
common UI in a popover accessed in the address bar, or a side
pane in the browser itself.
If a user logs in or does something requiring cookies that would
otherwise prevent normal functionality, prompt them with a
Permissions box if they haven't already accepted it in the usual
(optional) UI.
YetAnotherNick wrote 19 hours 16 min ago:
There isn't any way EU didn't knew this was possible and is a
better choice. There already was DNT header that they can
regulate. It also knew the harm to ad industry.
Fraaaank wrote 18 hours 49 min ago:
There isn't any rule that requires websites to use a cookie
banner. Your required to obtain explicit consent before
reading/setting any cookies that aren't strictly necessary.
The web came up with the cookie banner.
Google could've implemented a consent API in Chrome, but they
didn't. Guess why.
kalaksi wrote 23 hours 24 min ago:
Cookies for normal functionality don't require consent anyway.
But yes, I think just about everybody would like the UX you
described. But the entities that track you don't want to make
it that easy. You probably know of the do-not-track header too.
bazmattaz wrote 1 day ago:
This is so cool thanks for sharing. I can imagine itâs not
technically possible (yet?) but it would be cool if this could simply
be run as a browser extension rather than running a docker container
MarioMan wrote 1 day ago:
There are a couple of WebGPU LLM platforms available that form the
building blocks to accomplish this right from the browser,
especially since the models are so small. [1] [2] You do have to
worry about WebGPU compatibility in browsers though.
URI [1]: https://github.com/mlc-ai/web-llm
URI [2]: https://huggingface.co/docs/transformers.js/en/index
URI [3]: https://caniuse.com/webgpu
throwup238 wrote 1 day ago:
It should be possible using native messaging [1] which can call out
to an external binary. The 1password extensions use that to
communicate with the password manager binary.
URI [1]: https://developer.mozilla.org/en-US/docs/Mozilla/Add-ons/W...
antonok wrote 1 day ago:
I did actually make a rough proof-of-concept of this! One of my
long-term visions is to have it running natively in-browser, and
able to automatically fix site issues caused by adblocking whenever
they happen.
The PoC is a bit outdated but it's here:
URI [1]: https://github.com/brave/cookiemonster/tree/webext
binarysneaker wrote 1 day ago:
Maybe it could also send automated petitions to the EU to undo cookie
consent legislation, and reverse some of the enshitification.
sebastiennight wrote 1 day ago:
To me this take is like smokers complaining that the evil
government is forcing the good tobacco companies to degrade the
experience by adding pictures of cancer patients on cigarette
packs.
kortilla wrote 1 day ago:
Those donât really work:
URI [1]: https://jamanetwork.com/journals/jamanetworkopen/fullart...
shiftingleft wrote 1 day ago:
Do they help deter people from becoming smokers in the first
place?
kortilla wrote 11 hours 9 min ago:
Not sure if much serious research has been put into it. I
would be suspicious of it deterring them because a lot of
initial smoking happens in social situations where friends
pass out individual cigarettes.
By the time someone buys their own pack they are probably
hooked.
I suspect the obscene taxes blocking out young folks is one
of the most effective strategies
K0balt wrote 1 day ago:
I think there is real potential here, for smart browsing. Have the
llm get the page, replace all the ads with kittens, find
non-paywall versions if possible and needed, spoof fingerprint
data, detect and highlight AI generated drivel, etc. The site would
have no way of knowing that it wasnât touching eyeballs. We
might be able to rake back a bit of the web this way.
antonok wrote 1 day ago:
You probably wouldn't want to run this in real-time on every site
as it'll significantly increase the load on your browser, but as
long as it's possible to generate adblock filter rules, the fixes
can scale to a pretty large audience.
Tepix wrote 23 hours 20 min ago:
Depends on your machine and on the LLM. Could be doable.
K0balt wrote 1 day ago:
I was thinking running it in my home lab server as a proxy, but
yeah, scaling it to the browser would require some pretty
strong hardware. Still, maybe in a couple of years it could be
mainstream.
antonok wrote 1 day ago:
Ha, I'm not sure the EU is prepared to handle the deluge of
petitions that would ensue.
On a more serious note, this must be the first time we can
quantitatively measure the impact of cookie consent legislation
across the web, so maybe there's something to be explored there.
pk-protect-ai wrote 1 day ago:
why don't you spam the companies who want your data instead? The
sites can simply stop gathering your data, then they will not
require to ask for consent ...
whywhywhywhy wrote 1 day ago:
Because they have no reason to care about what you think or
feel or they wouldn't be doing it in the first place.
Cookie notices just gave them another weapon in the end.
frail_figure wrote 1 day ago:
Itâs the same comments on HN as always. They think EU setting
up rules is somehow worse than companies breaking them. We see
how the US is turning out without pesky EU restrictions :)
GardenLetter27 wrote 1 day ago:
The US has 3x higher salaries, larger houses and a much
higher quality of life?
I work as a senior engineer in Europe and make barely $4k net
per month... and that's considered a "good" salary!
Lutger wrote 1 day ago:
It has higher salaries for privileged people like senior
engineers. Try making ends meet in a lower class job.
And you have (almost) free and universal healthcare in
Europa, good food available everywhere, drinking water that
doesn't poison you, walkable cities, good public transport,
somewhat decent police and a functioning legal system. The
list goes on. Does this not impact your quality of life? Do
you not care about these things?
How can you have a higher quality of life as a society with
higher murders, much lower life-expectancy, so many people
in jail, in debt, etc.
macinjosh wrote 1 day ago:
Touch grass. The US is a big place and is nothing like
you seem to think it is.
Europe on the other hand can't even manage to defend
itself and relies on the US for their sheer existence.
pona-a wrote 22 hours 54 min ago:
Can you enlighten me of a state where none of parent's
points apply? I'd be glad to be educated.
thetrash wrote 1 day ago:
I programmed my own version of Tic Tac Toe in Godot, using a Llama 3B
as the AI opponent. Not for work flow, but figuring out how to beat it
is entertaining during moments of boredom.
spiritplumber wrote 1 day ago:
Number of players: zero
U.S. FIRST STRIKE WINNER: NONE
USSR FIRST STRIKE WINNER: NONE
NATO / WARSAW PACT WINNER: NONE
FAR EAST STRATEGY WINNER: NONE
US USSR ESCALATION WINNER: NONE
MIDDLE EAST WAR WINNER: NONE
USSR CHINA ATTACK WINNER: NONE
INDIA PAKISTAN WAR WINNER: NONE
MEDITERRANEAN WAR WINNER: NONE
HONGKONG VARIANT WINNER: NONE
Strange game. The only winning move is not to play
cwmoore wrote 1 day ago:
I'm playing with the idea of identifying logical fallacies stated by
live broadcasters.
halJordan wrote 16 hours 57 min ago:
Logical fallacies are oftentimes totally relevant during anything
that is not predicate logic. I'm not wrong for saying "The Surgeon
General says smoking is bad, you shouldn't smoke." That's a perfectly
reasonable appeal to authority.
genewitch wrote 15 hours 30 min ago:
It's still a fallacy, though. I hope we can agree on that part. If
you have something map-reducing audio to timestamps of fallacies by
who said them it makes it gamified and you can use the information
shown to decide how much weight to give to their words.
thesz wrote 17 hours 52 min ago:
I think this is the best idea thus far!
Keep good work, good fellow. ;)
grisaitis wrote 1 day ago:
even better, podcasters probably easier to fetch the data as well
vaylian wrote 1 day ago:
LLMs are notoriously unreliable with mathematics and logic. I wish
you the best of luck, because this would nevertheless be an awesome
tool to have.
JayStavis wrote 1 day ago:
Automation to identify logical/rhetorical fallacies is a long held
dream of mine, would love to follow along with this project if it
picks up somehow
petesergeant wrote 1 day ago:
I'll be very positively impressed if you make this work; I spend all
day every day for work trying to make more capable models perform
basic reasoning, and often failing :-P
genewitch wrote 1 day ago:
I have several rhetoric and logic books of the sort you might use for
training or whatever, and one of my best friends got a doctorate in
a tangential field, and may have materials and insights.
We actually just threw a relationship curative app online in 17 hours
around Thanksgiving., so they "owe" me, as it were.
I'm one of those people that can do anything practical with tech and
the like, but I have no imagination for it - so when someone mentions
something that I think would be beneficial for my fellow humans I get
this immense desire to at least cheer on if not ask to help.
spiritplumber wrote 1 day ago:
That's fantastic and I'd love to help
cwmoore wrote 1 day ago:
So far not much beyond this list of targets to identify
URI [1]: https://en.wikipedia.org/wiki/List_of_fallacies
nozzlegear wrote 1 day ago:
I have a small fish script I use to prompt a model to generate three
commit messages based off of my current git diff. I'm still playing
around with which model comes up with the best messages, but usually I
only use it to give me some ideas when my brain isn't working. All the
models accomplish that task pretty well.
Here's the script: [1] And for this change [1] it generated these
messages:
1. `fix: change from printf to echo for handling git diff input`
2. `refactor: update codeblock syntax in commit message generator`
3. `style: improve readability by adjusting prompt formatting` [1]
URI [1]: https://github.com/nozzlegear/dotfiles/blob/master/fish-functi...
URI [2]: https://github.com/nozzlegear/dotfiles/commit/0db65054524d0d2e...
mystified5016 wrote 21 hours 52 min ago:
That's actually pretty useful. This could be a big help in betting
back into the groove when you leave uncommitted changes over the
weekend.
A summary of changes like this might be just enough to spark your
memory on what you were actually doing with the changes. I'll have to
give it a shot!
lionkor wrote 1 day ago:
Those commit messages are pretty terrible, please try to come up with
actual messages ;)
relistan wrote 1 day ago:
Interesting idea. But those say whatâs in the commit. The commit
diff already tells you that. The best commit messages IMO tell you
why you did it and what value was delivered. I think itâs gonna be
hard for an LLM to do that since that context lives outside the code.
But maybe it would, if you hook it to e.g. a ticketing system and
include relevant tickets so it can grab context.
For instance, in your first example, why was that change needed? It
was a fix, but for what issue?
In the second message: why was that a desirable change?
zanderwohl wrote 23 hours 41 min ago:
> The commit diff already tells you that.
When you squash a branch you'll have 200+ lines of new code on a
new feature. The diff is not a quick way to get a summary of what's
happening. You should put the "what" in your commit messages.
nozzlegear wrote 1 day ago:
Typically I put the "why" of the commit in the body unless it's a
super simple change, but that's a good point. Sometimes this
function does generate a commit body to go with the summary, and
sometimes it doesn't. It also has a habit of only looking at the
first file in a diff and basing its messages off of that, instead
of considering the whole patch.
I'll tweak the prompt when I have some time today and see if I can
get some more consistency out of it.
rane wrote 1 day ago:
Most of the time you are not able to fit the "Why?" in the summary.
That's what the body of the commit message is for.
lnenad wrote 1 day ago:
I disagree. When you look at the git history in x months you're
gonna have a hard time understanding what was done following your
example.
Draiken wrote 1 day ago:
I disagree. If you look back and all you see are commit messages
summarizing the diff, you won't get any meaningful information.
Telling me `Changed timeout from 30s to 60s` means nothing, while
`Increase timeout for slow requests` gives me an actual idea of
why that was done.
Even better if you add meaningful messages to the commit body.
Take a look at commits from large repositories like the Linux
kernel and we can see how good commit messages looks like.
lnenad wrote 6 hours 59 min ago:
I mean you're not op but his comment was saying
> Interesting idea. But those say whatâs in the commit. The
commit diff already tells you that. The best commit messages
IMO tell you why you did it and what value was delivered.
Which doesn't include what was done. Your example includes both
which is fine. But not including what the commit does in the
message is an antipattern imho. Everything else that is added
is a bonus.
Draiken wrote 5 hours 11 min ago:
Many changes require multiple smaller changes, so this is not
always possible.
For me the commit message should tell me the what/why and the
diff is the how. It's great to understand if, for example, a
change was intentional or a bug.
Many times when searching for the source of a bug I could not
tell if the line changed was intentional or a mistake because
the commit message was simply repeating what was on the diff.
If you say your intention was to add something and the diff
shows a subtraction, you can easily tell it was a mistake.
Contrived example but I think it demonstrates my point.
This only really works if commits are meaningful though. Most
people are careless and half their commits are 'fix this',
'fix again', 'wip', etc. At that point the only place that
can contain useful information on the intentions are the pull
requests/issues around it.
Take a single commit from the Linux kernel: [1] It doesn't
tell me "add function X, Y and boolean flag Z". It tells us
what/why it was done, and the diff shows us how.
URI [1]: https://github.com/torvalds/linux/commit/08bd5b7c9a2...
relistan wrote 1 day ago:
By adding more context? Iâm not sure who youâre replying to
or what your objection is.
mentos wrote 1 day ago:
Awesome need to make one for naming variables too haha
deivid wrote 1 day ago:
Not sure it qualifies, but I've started building an Android app that
wraps bergamot[0] (the firefox translation models) to have on-device
translation without reliance on google.
Bergamot is already used inside firefox, but I wanted translation also
outside the browser.
[0]: bergamot
URI [1]: https://github.com/browsermt/bergamot-translator
deivid wrote 1 day ago:
I would be very interested if someone is aware of any small/tiny
models to perform OCR, so the app can translate pictures as well
Eisenstein wrote 1 day ago:
MiniCPM-V 2.6 isn't that small (8b) but it can do this.
Here is a demo.
* [1] Using this script:
*
URI [1]: https://i.imgur.com/pAuTeAf.jpeg
URI [2]: https://github.com/jabberjabberjabber/LLMOCR/
ata_aman wrote 1 day ago:
I have it running on a Raspberry Pi 5 for offline chat and RAG.
I wrote this open-source code for it: [1] It also does RAG on apps
there, like the music player, contacts app and to-do app. I can ask it
to recommend similar artists to listen to based on my music library for
example or ask it to quiz me on my PDF papers.
URI [1]: https://github.com/persys-ai/persys
nejsjsjsbsb wrote 1 day ago:
Does [1] run on the rpi?
Is that design 3d printable? Or is that for paid users only.
URI [1]: https://github.com/persys-ai/persys-server
ata_aman wrote 1 day ago:
I can publish it no problem. Iâll create a new repo with
instructions for the hardware with CAD files.
Designing a new one for the NVIDIA Orin Nano Super so it might take
a few days.
nejsjsjsbsb wrote 1 day ago:
Up to you! Totally understand if you want to hold something back
for a paid option!
kristopolous wrote 1 day ago:
I'm working on using them for agentic voice commands of a limited
scope.
My needs are narrow and limited but I want a bit of flexibility.
simonjgreen wrote 1 day ago:
Micro Wake Word is a library and set of on device models for ESPs to
wake on a spoken wake word. [1] Recently deployed in Home Assistants
fully local capable Alexa replacement.
URI [1]: https://github.com/kahrendt/microWakeWord
URI [2]: https://www.home-assistant.io/voice_control/about_wake_word/
yzydserd wrote 1 day ago:
Nice idea.
kortilla wrote 1 day ago:
Make sure your meeting participants know youâre transcribing
them. Has similar notification requirements as recording state to
state.
flippyhead wrote 1 day ago:
I have a tiny device that listens to conversations between two people
or more and constantly tries to declare a "winner"
TechDebtDevin wrote 1 day ago:
Your SO must really love that lmao
deivid wrote 1 day ago:
what model do you use for speech to text?
prakashn27 wrote 1 day ago:
wifey always wins. ;)
nejsjsjsbsb wrote 1 day ago:
All computation on device?
mkaic wrote 1 day ago:
This reminds me of the antics of streamer DougDoug, who often uses
LLM APIs to live-summarize, analyze, or interact with his (often
multi-thousand-strong) Twitch chat. Most recently I saw him do a
GeoGuessr stream where he had ChatGPT assume the role of a detective
who must comb through the thousands of chat messages for clues about
where the chat thinks the location is, then synthesizes the clamor
into a final guess. Aside from constantly being trolled by people
spamming nothing but "Kyoto, Japan" in chat, it occasionaly
demonstrated a pretty effective incarnation of "the wisdom of the
crowd" and was strikingly accurate at times.
eddd-ddde wrote 1 day ago:
I love that there's not even a vague idea of the winner "metric" in
your explanation. Like it's just, _the_ winner.
hn8726 wrote 1 day ago:
What approach/stack would you recommend for listening to an ongoing
conversation, transcribing it and passing through llm? I had some use
cases in mind but I'm not very familiar with AI frameworks and tools
jjcm wrote 1 day ago:
Are you raising a funding round? I'm bought in. This is hilarious.
amelius wrote 1 day ago:
You can use the model to generate winning speeches also.
econ wrote 1 day ago:
This is a product I want
oa335 wrote 1 day ago:
This made me actually laugh out loud. Can you share more details on
hardware and models used?
pseudosavant wrote 1 day ago:
I'd love to hear more about the hardware behind this project. I've
had concepts for tech requiring a mic on me at all times for various
reasons. Always tricky to have enough power in a reasonable DIY form
factor.
mritchie712 wrote 1 day ago:
I used local LLMs via Ollama for generating H1's / marketing copy.
1. Create several different personas
2. Generate a ton of variation using a high temperature
3. Compare the variagtions head-to-head using the LLM to get a win /
loss ratio
The best ones can be quite good.
0 -
URI [1]: https://www.definite.app/blog/overkillm
Mashimo wrote 1 day ago:
What is an H1?
laristine wrote 1 day ago:
Main heading of an article
TachyonicBytes wrote 1 day ago:
Not the OP, but they are "Headers". Probably coming from the tag
in html. What outsiders probably call "Headlines".
UltraSane wrote 1 day ago:
clever name!
ignoramous wrote 1 day ago:
We're prototyping a text firewall (for Android) with Gemma2 2B (which
limits us to English), though DeepSeek's R1 variants now look pretty
promising [0]: Depending on the content, we rewrite the text or
quarantine it from your view. Of course this is easy (for English) in
the sense that the core logic is all LLMs [1], but the integration
points (on Android) are not so straight forward for anything other than
SMS. [2] A more difficult problem we forsee is to turn it into a
real-time (online) firewall (for calls, for example). [1] MediaPipe in
particular makes it simple to prototype around Gemma2 on Android: [2]
Intend to open source it once we get it working for anything other than
SMSes
URI [1]: https://chat.deepseek.com/a/chat/s/d5aeeda1-fefe-4fc6-8c90-20e...
URI [2]: https://ai.google.dev/edge/mediapipe/solutions/genai/llm_infer...
deet wrote 1 day ago:
We (avy.ai) are using models in that range to analyze computer activity
on-device, in a privacy sensitive way, to help knowledge workers as
they go about their day.
The local models do things ranging from cleaning up OCR, to summarizing
meetings, to estimating the user's current goals and activity, to
predicting search terms, to predicting queries and actions that, if
run, would help the user accomplish their current task.
The capabilities of these tiny models have really surged recently. Even
small vision models are becoming useful, especially if fine tuned.
bendews wrote 1 day ago:
Is this along the lines of rewind.ai, MSCopilot, screenpipe, or
something else entirely?
A4ET8a8uTh0_v2 wrote 1 day ago:
Kinda? All local so very much personal, non-business use. I made Ollama
talk in a specific persona styles with the idea of speaking like Spider
Jerusalem, when I feel like retaining some level of privacy by avoiding
phrases I would normally use. Uncensored llama just rewrites my post
with a specific persona's 'voice'. Works amusingly well for that
purpose.
eb0la wrote 1 day ago:
We're using small language models to detect prompt injection. Not too
cool, but at least we can publish some AI-related stuff on the internet
without a huge bill.
sitkack wrote 1 day ago:
What kind of prompt injection attacks do you filter out? Have you
tested with a prompt tuning framework?
behohippy wrote 1 day ago:
I have a mini PC with an n100 CPU connected to a small 7" monitor
sitting on my desk, under the regular PC. I have llama 3b (q4)
generating endless stories in different genres and styles. It's fun to
glance over at it and read whatever it's in the middle of making. I
gave llama.cpp one CPU core and it generates slow enough to just read
at a normal pace, and the CPU fans don't go nuts. Totally not
productive or really useful but I like it.
droideqa wrote 1 day ago:
That's awesome!
ipython wrote 1 day ago:
That's neat. I just tried something similar:
FORTUNE=$(fortune) && echo $FORTUNE && echo "Convert the
following output of the Unix `fortune` command into a small
screenplay in the style of Shakespeare: \n\n $FORTUNE" | ollama run
phi4
watermelon0 wrote 1 day ago:
Doesn't `fortune` inside double quotes execute the command in bash?
You should use single quotes instead of backticks.
keeganpoppen wrote 1 day ago:
oh wow that is actually such a brilliant little use case-- really
cuts to the core of the real "magic" of ai: that it can just keep
running continuously. it never gets tired, and never gets tired of
thinking.
Uehreka wrote 1 day ago:
Do you find that it actually generates varied and diverse stories? Or
does it just fall into the same 3 grooves?
Last week I tried to get an LLM (one of the recent Llama models
running through Groq, it was 70B I believe) to produce randomly
generated prompts in a variety of styles and it kept producing
cyberpunk scifi stuff. When I told it to stop doing cyberpunk scifi
stuff it went completely to wild west.
jaggs wrote 20 hours 27 min ago:
URI [1]: https://old.reddit.com/r/LocalLLaMA/comments/1i615u1/the_f...
greenavocado wrote 1 day ago:
Set temperature to 1.0
behohippy wrote 1 day ago:
It's a 3b model so the creativity is pretty limited. What helped
for me was prompting for specific stories in specific styles. I
have a python script that randomizes the prompt and the writing
style, including asking for specific author styles.
TMWNN wrote 1 day ago:
> Do you find that it actually generates varied and diverse
stories? Or does it just fall into the same 3 grooves?
> Last week I tried to get an LLM (one of the recent Llama models
running through Groq, it was 70B I believe) to produce randomly
generated prompts in a variety of styles and it kept producing
cyberpunk scifi stuff.
100% relevant: "Someday" < [1] > by Isaac Asimov, 1956
URI [1]: https://en.wikipedia.org/wiki/Someday_(short_story)
coder543 wrote 1 day ago:
Someone mentioned generating millions of (very short) stories with
an LLM a few weeks ago: [1] They linked to an interactive explorer
that nicely shows the diversity of the dataset, and the HF repo
links to the GitHub repo that has the code that generated the
stories: [2] So, it seems there are ways to get varied stories.
URI [1]: https://news.ycombinator.com/item?id=42577644
URI [2]: https://github.com/lennart-finke/simple_stories_generate
janalsncm wrote 1 day ago:
Generate a list of 5000 possible topics youâd like it to talk
about. Randomly pick one and inject that into your prompt.
o11c wrote 1 day ago:
You should not ever expect an LLM to actually do what you want
without handholding, and randomness in particular is one of the
places it fails badly. This is probably fundamental.
That said, this is also not helped by the fact that all of the
default interfaces lack many essential features, so you have to
build the interface yourself. Neither "clear the context on every
attempt" nor "reuse the context repeatedly" will give good results,
but having one context producing just one-line summaries, then
fresh contexts expanding each one will do slightly less badly.
(If you actually want the LLM to do something useful, there are
many more things that need to be added beyond this)
dotancohen wrote 1 day ago:
Sounds to me like you might want to reduce the Top P - that will
prevent the really unlikely next tokens from ever being selected,
while still providing nice randomness in the remaining next
tokens so you continue to get diverse stories.
bithavoc wrote 1 day ago:
this is so cool, any chance you post a video?
behohippy wrote 1 day ago:
Just this pic:
URI [1]: https://imgur.com/ip8GWIh
Dansvidania wrote 1 day ago:
this sounds pretty cool, do you have any video/media of it?
behohippy wrote 1 day ago:
I don't have a video but here's a pic of the output:
URI [1]: https://imgur.com/ip8GWIh
sky2224 wrote 13 hours 59 min ago:
The next step is to format it so it looks like an endless
starwars intro.
azhenley wrote 1 day ago:
Microsoft published a paper on their FLAME model (60M parameters) for
Excel formula repair/completion which outperformed much larger models
(>100B parameters).
URI [1]: https://arxiv.org/abs/2301.13779
coder543 wrote 1 day ago:
That paper is from over a year ago, and it compared against
codex-davinci... which was basically GPT-3, from what I understand.
Saying >100B makes it sound a lot more impressive than it is in
today's context... 100B models today are a lot more capable. The
researchers also compared against a couple of other
ancient(/irrelevant today), small models that don't give me much
insight.
FLAME seems like a fun little model, and 60M is truly tiny compared
to other LLMs, but I have no idea how good it is in today's context,
and it doesn't seem like they ever released it.
aDyslecticCrow wrote 21 hours 28 min ago:
I would like to disagree with its being irrelevant. If anything,
the 100B models are irrelevant in the context and should be seen as
a "fun inclusion" rather than a serious addition worth comparing
against. It out-performing a 100B model at the time becomes a fun
bragging point, but it's not the core value of the method or paper.
Running a prompt against every single cell of a 10k row document
was never gonna happen with a large model. Even using a transformer
model architecture in the first place can be seen as ludicrous
overkill but feasible on modern machines.
So I'd say the paper is very relevant, and the top commenter in
this very thread demonstrated their own homegrown version with a
very nice use-case (paper abstract and title sorting for making a
summary paper)
coder543 wrote 21 hours 23 min ago:
> Running a prompt against every single cell of a 10k row
document was never gonna happen with a large model
That isnât the main point of FLAME, as I understood it. The
main point was to help you when youâre editing a particular
cell. codex-davinci was used for real time Copilot tab
completions for a long time, I believe, and editing within a
single formula in a spreadsheet is far less demanding than
editing code in a large document.
After I posted my original comment, I realized I should have
pointed out that Iâm fairly sure we have 8B models that handily
outperform codex-davinci these days⦠further driving home how
irrelevant the claim of â>100Bâ was here (not talking about
the paper). Plus, an off the shelf model like Qwen2.5-0.5B (a
494M model) could probably be fine tuned to compete with (or
dominate) FLAME if you had access to the FLAME training data â
there is probably no need to train a model from scratch, and a
0.5B model can easily run on any computer that can run the
current version of Excel.
You may disagree, but my point was that claiming a 60M model
outperforms a 100B model just means something entirely different
today. Putting that in the original comment higher in the thread
creates confusion, not clarity, since the models in question are
very bad compared to what exists now. No one had clarified that
the paper was over a year old until I commented⦠and FLAME was
being tested against models that seemed to be over a year old
even when the paper was published. I donât understand why the
researchers were testing against such old models even back then.
3abiton wrote 1 day ago:
But I feel we're going back full circle. These small models are not
generalist, thus not really LLMs at least in terms of objective.
Recently there has been a rise of "specialized" models that provide
lots of values, but that's not why we were sold on LLMs.
janalsncm wrote 1 day ago:
I think playing word games about what really counts as an LLM is a
losing battle. It has become a marketing term, mostly. Itâs
better to have a functionalist point of view of âwhat can this
thing doâ.
Suppafly wrote 1 day ago:
Specialized models work much better still for most stuff. Really we
need an LLM to understand the input and then hand it off to a
specialized model that actually provides good results.
colechristensen wrote 1 day ago:
But that's the thing, I don't need my ML model to be able to write
me a sonnet about the history of beets, especially if I want to run
it at home for specific tasks like as a programming assistant.
I'm fine with and prefer specialist models in most cases.
zeroCalories wrote 1 day ago:
I would love a model that knows SQL really well so I don't need
to remember all the small details of the language. Beyond that, I
don't see why the transformer architecture can't be applied to
any problem that needs to predict sequences.
dr_kiszonka wrote 1 day ago:
The trick is to find such problems with enough training data
and some market potential. I am terrible at it.
andai wrote 1 day ago:
This is wild. They claim it was trained exclusively on Excel
formulas, but then they mention retrieval? Is it understanding the
connection between English and formulas? Or am I misunderstanding
retrieval in this context?
Edit: No, the retrieval is Formula-Formula, the model (nor I believe
tokenizer) does not handle English.
barrenko wrote 1 day ago:
This is really cool. Is this already in Excel?
arionhardison wrote 1 day ago:
I am, in a way by using EHR/EMR data for fine tuning so agents can
query each other for medical records in a HIPPA compliant manner.
Havoc wrote 1 day ago:
Pretty sure they are mostly used as fine tuning targets, rather than
as-is.
dcl wrote 1 day ago:
But for what purposes?
iamnotagenius wrote 1 day ago:
No, but I use llama 3.2 1b and qwen2.5 1.5 as bash oneliner generator,
always runnimg in console.
XMasterrrr wrote 1 day ago:
What's your workflow like? I use AI Chat. I load
Qwen2.5-1.5B-Instruct with llama.cpp server, fully offloaded to the
CPU, and then I config AI Chat to connect to the llama.cpp endpoint.
andai wrote 1 day ago:
Could you elaborate?
iamnotagenius wrote 1 day ago:
I just run llama-cli with the model. Every time I want some "awk"
or "find" trickery, I just ask model. Good for throwaway python
scripts too.
jajko wrote 1 day ago:
Can it do 'sed'?
I think one major improvement for folks like me would be
human->regex LLM translator, ideally also respecting different
flavors/syntax for various languages and tools.
This has been a bane of me - I run into requirement to develop
some complex regexes maybe every 2-3 years, so I dig deep into
specs, work on it, deliver eventually if its even possible, and
within few months almost completely forget all the details and
start at almost same place next time. It gets better over time
but clearly I will retire earlier than this skill settles in
well.
iamnotagenius wrote 22 hours 44 min ago:
have not tried yet. any specific query? I can try.
XMasterrrr wrote 1 day ago:
I think I know what he means. I use AI Chat. I load
Qwen2.5-1.5B-Instruct with llama.cpp server, fully offloaded to the
CPU, and then I config AI Chat to connect to the llama.cpp
endpoint.
Checkout the demo they have below
URI [1]: https://github.com/sigoden/aichat#shell-assistant
RhysU wrote 1 day ago:
"Comedy Writing With Small Generative Models" by Jamie Brew (Strange
Loop 2023) [1] Spend the 45 minutes watching this talk. It is a
delight. If you are unsure, wait until the speaker picks up the guitar.
URI [1]: https://m.youtube.com/watch?v=M2o4f_2L0No
prettyblocks wrote 16 hours 54 min ago:
Excellent share - nice to see people doing cool things with the tech
while not taking themselves too seriously.
100k wrote 1 day ago:
Seconded! This was my favorite talk at Strange Loop (including my
own).
mettamage wrote 1 day ago:
I simply use it to de-anonymize code that I typed in via Claude
Maybe should write a plugin for it (open source):
1. Put in all your work related questions in the plugin, an LLM will
make it as an abstract question for you to preview and send it
2. And then get the answer with all the data back
E.g. df[âcookie_company_nameâ] becomes df[âaâ] and back
sundarurfriend wrote 1 day ago:
You're using it to anonymize your code, not de-anonymize someone's
code. I was confused by your comment until I read the replies and
realized that's what you meant to say.
kreyenborgi wrote 1 day ago:
I read it the other way, their code contains eg fetch(url,
pw:hunter123), and they're asking Claude anonymized questions like
"implement handler for fetch(url, {pw:mycleartrxtpw})"
And then claude replies
fetch(url, {pw:mycleartrxtpw}).then(writething)
And then the local llm converts the placeholder mycleartrxtpw into
hunter123 using its access to the real code
mettamage wrote 1 day ago:
It's that yea
Flow would be:
1. Llama prompt: write a console log statement with my username
and password: mettamage, superdupersecret
2. Claude prompt (edited by Llama): write a console log statement
with my username and password: asdfhjk, sdjkfa
3. Claude replies: console.log('asdfhjk', 'sdjkfa')
4. Llama gets that input and replies to me:
console.log('mettamage', 'superdupersecret')
sundarurfriend wrote 1 day ago:
> Put in all your work related questions in the plugin, an LLM
will make it as an abstract question for you to preview and send
it
So the LLM does both the anonymization into placeholders and then
later the replacing of the placeholders too. Calling the latter
step de-anonymization is confusing though, it's "de-anonymizing"
yourself to yourself. And the overall purpose of the plugin is to
anonymize OP to Claude, so to me at least that makes the whole
thing clearer.
mettamage wrote 1 day ago:
I could've been a bit more clear, sorry about that.
sauwan wrote 1 day ago:
Are you using the model to create a key-value pair to find/replace
and then reverse to reanonymize, or are you using its outputs
directly? If the latter, is it fast enough and reliable enough?
sitkack wrote 1 day ago:
So you are using a local small model to remove identifying
information and make the question generic, which is then sent to a
larger model? Is that understanding correct?
I think this would have some additional benefits of not confusing the
larger model with facts it doesn't need to know about. My erasing
information, you can allow its attention heads to focus on the pieces
that matter.
Requires further study.
mettamage wrote 1 day ago:
> So you are using a local small model to remove identifying
information and make the question generic, which is then sent to a
larger model? Is that understanding correct?
Yep that's it
politelemon wrote 1 day ago:
Could you recommend a tiny language model I could try out locally?
mettamage wrote 1 day ago:
Llama 3.2 has about 3.2b parameters. I have to admit, I use bigger
ones like phi-4 (14.7b) and Llama 3.3 (70.6b) but I think Llama 3.2
could do de-anonimization and anonimization of code
RicoElectrico wrote 1 day ago:
Llama 3.2 punches way above its weight. For general "language
manipulation" tasks it's good enough - and it can be used on a
CPU with acceptable speed.
seunosewa wrote 1 day ago:
How many tokens/s?
iamnotagenius wrote 1 day ago:
10-15t/s on 12400 with ddr5
OxfordOutlander wrote 1 day ago:
+1 this idea. I do the same. Just do it locally using ollama,
also using 3.2 3b
psyklic wrote 1 day ago:
JetBrains' local single-line autocomplete model is 0.1B (w/ 1536-token
context, ~170 lines of code): [1] For context, GPT-2-small is 0.124B
params (w/ 1024-token context).
URI [1]: https://blog.jetbrains.com/blog/2024/04/04/full-line-code-comp...
staticautomatic wrote 1 day ago:
Is that why their tab completion is so bad now?
sam_lowry_ wrote 1 day ago:
Hm... I wonder what your use case it. I do the modern Enterprise
Java and the tab completion is a major time saver.
While interactive AI is all about posing, meditating on the prompt,
then trying to fix the outcome, IntelliJ tab completion... shows
what it will complete as you type and you Tab when you are 100% OK
with the completion, which surprisingly happens 90..99% of the time
for me, depending on the project.
pseudosavant wrote 1 day ago:
I wonder how big that model is in RAM/disk. I use LLMs for FFMPEG all
the time, and I was thinking about training a model on just the
FFMPEG CLI arguments. If it was small enough, it could be a package
for FFMPEG. e.g. `ffmpeg llm "Convert this MP4 into the latest
royalty-free codecs in an MKV."`
binary132 wrote 1 day ago:
Thatâs a great idea, but I feel like it might be hard to get it
to be correct enough
maujim wrote 1 day ago:
from a few days ago:
URI [1]: https://news.ycombinator.com/item?id=42706637
h0l0cube wrote 1 day ago:
Please submit a blog post to HN when you're done. I'd be curious
to know the most minimal LLM setup needed get consistently sane
output for FFMPEG parameters.
jedbrooke wrote 1 day ago:
the jetbrains models are about 70MB zipped on disk (one model per
language)
pseudosavant wrote 23 hours 17 min ago:
That is easily small enough to host as a static SPA web app. I
was first thinking it would be cool to make a static web app that
would run the model locally. You'd make a query and it'd give the
FFMPEG commands.
smaddox wrote 1 day ago:
You can train that size of a model on ~1 billion tokens in ~3 minutes
on a rented 8xH100 80GB node (~$9/hr on Lambda Labs, RunPod io, etc.)
using the NanoGPT speed run repo: [1] For that short of a run, you'll
spend more time waiting for the node to come up, downloading the
dataset, and compiling the model, though.
URI [1]: https://github.com/KellerJordan/modded-nanogpt
WithinReason wrote 1 day ago:
That size is on the edge of something you can train at home
Sohcahtoa82 wrote 1 day ago:
Not even on the edge. That's something you could train on a 2 GB
GPU.
The general guidance I've used is that to train a model, you need
an amount of RAM (or VRAM) equal to 8x the number of parameters, so
a 0.125B model would need 1 GB of RAM to train.
vineyardmike wrote 1 day ago:
If you have modern hardware, you can absolutely train that at home.
Or very affordable on a cloud service.
Iâve seen a number of âDIY GPT-2â tutorials that target this
sweet spot. You wonât get amazing results unless you want to
leave a personal computer running for a number of hours/days and
you have solid data to train on locally, but fine-tuning should be
in the realm of normal hobbyists patience.
nottorp wrote 1 day ago:
Hmm is there anything reasonably ready made* for this spot?
Training and querying a llm locally on an existing codebase?
* I don't mind compiling it myself but i'd rather not write it.
DIR <- back to front page