frontpage.
newsnewestaskshowjobs

Open Source @Github

fp.

507 Mechanical Movements

https://507movements.com/
270•helloplanets•3h ago•35 comments

Small Models Have Arrived

https://calv.info/small-models-have-arrived
103•tosh•1h ago•36 comments

Saving 100 terabytes of memory by optimizing 1.1.1.1's DNS cache

https://blog.cloudflare.com/dns-cache-memory-optimization-1111/
22•TangerineDream•19m ago•0 comments

Suica, Japan's First IC Transit Card

https://www.tokyodev.com/articles/the-story-of-suica
53•zdw•1h ago•34 comments

Nvidia projects $673B in sales as AI demand widens

https://forgeeks.net/nvidia-673-billion-ai-growth-forecast/
63•kuuuzya•2h ago•35 comments

Microduck

https://pollen-robotics.com/microduck/
322•robotswantdata•6h ago•127 comments

Bild AI (YC W25) Is Hiring Product and AI Engineers

https://www.bild.ai/jobs
1•rooppal•37m ago

Emacs 31: An unofficial guide to Markdown-ts-mode

https://rahuljuliato.com/posts/markdown-ts-mode-emacs-31
106•RahulMJ•4h ago•39 comments

Decompiling a Nintendo 64 Game in 84 Days

https://blog.chrislewis.au/decompiling-a-nintendo-64-game-in-84-days/
46•knackers•2h ago•10 comments

CoMaps integration with the wider FLOSS ecosystem

https://www.comaps.app/news/2026-08-23/comaps-integration-with-the-wider-floss-ecosystem/
104•janandonly•4h ago•12 comments

Launch HN: Salem Robotics (YC S26) – Software for industrial inspection robots

22•Salem_robotics•1h ago•15 comments

Launching Route 53 Files

https://www.daemonology.net/blog/2026-08-27-Launching-Route-53-Files.html
49•louis-paul•2h ago•18 comments

Show HN: My Claude quota ran out in 10 minutes, so I made a tool to find out why

https://github.com/kelviq/tare
15•sachinneravath•59m ago•1 comments

Yayoi Kusama has died

https://www.bbc.com/news/articles/c3v4k0re3vwo
99•herbertl•1h ago•11 comments

MIT's Ad Hoc Committee on AI Use in Teaching, Learning, and Research Training

https://aiandeducation.mit.edu/report/
72•pbui•4h ago•45 comments

A curmudgeon tries a language server

https://entropicthoughts.com/curmudgeon-tries-language-server
79•crescit_eundo•1d ago•52 comments

Why HPSC Is a Big Deal for Space Exploration

https://www.windriver.com/blog/Why-HPSC-Is-a-Big-Deal-for-Space-Exploration
11•mooreds•1h ago•3 comments

The turbulent AI era is here

https://www.gatesnotes.com/work/make-ai-work-for-everyone/reader/a-turbulent-ai-era-and-critical-...
24•nanna•1d ago•278 comments

The Teaser Period: Why the AI Boom Is Hitting a Reset Wall

https://www.groundbrkr.com/p/the-teaser-period-why-the-ai-boom
57•gtzi•3h ago•52 comments

Aphantasia Beginner's Guide

https://aphantasia.com/guide
51•ksec•4h ago•95 comments

Trade (and Tariffs)

https://xkcd.com/3290/
285•throw0101d•3h ago•103 comments

Nvidia agrees to acquire Hugging Face for $13B

https://www.businessinsider.com/nvidia-in-talks-to-buy-hugging-face-13-billion-dollars-2026-8
1703•mfiguiere•16h ago•795 comments

Grok Bot for Linux: Unofficial port of the official app (open source)

https://github.com/jakob-bu/grok-bot-linux-unofficial
19•j-bu•47m ago•3 comments

BRIN is 1/4570th the size of a B-tree, until 5% of rows are updated

https://deepsql.ai/blog/the-zonemap-shaped-hole-in-postgres
8•venkat971•1h ago•0 comments

Hollywood's Video Game Era Is Here. Will It Be Any Good?

https://www.statsignificant.com/p/hollywoods-video-game-era-is-here
19•NomNew•2h ago•30 comments

Show HN: Restoredrill – proves your Postgres backups restore

https://github.com/ahmadpiran/restoredrill
28•ahmadpiran•3h ago•9 comments

Engineered yeast for converting plastic and biomass compounds into food

https://acs.digitellinc.com/live/37/session/586399
13•ehwa37•1h ago•12 comments

Show HN: See fiber breaks linked to a map

https://react-networks-lib.rackout.net/otdr-strip
8•matt-p•59m ago•0 comments

Humanity has the debate about AI consciousness backwards

https://economist.com/by-invitation/2026/08/20/humanity-has-the-debate-about-ai-consciousness-bac...
31•semiquaver•15h ago•77 comments

CMS with AI, Not AI CMS: Wagtail 8.0's New API

https://wagtail.org/blog/cms-with-ai-not-ai-cms-wagtail-80s-new-api/
8•ThibWeb•1h ago•1 comments
Open in hackernews

Small Models Have Arrived

https://calv.info/small-models-have-arrived
98•tosh•1h ago

Comments

glimshe•1h ago
> There's obviously a lot we can optimize here, but if you're charging what the WSJ or The Economist charges, you'd better be delivering similar value.

Gosh, watching paint dry has been a better value than reading The Economist in the last 5 years or so.

That aside, I had good results with Luna. I'd be interested in hearing about a comparison that takes into consideration response time (not TPS), cost and performance of the popular models at different settings. That chart has some of that. For instance, is Luna Max a better value than Terra Medium?

yousif_123123•51m ago
Have you personally read it for 5 years to determine this? If not, how could you possibly hold this position?
swiftcoder•59m ago
I find it quite funny all these folks who are addicted to chasing frontier models, only just noticing that small models became "good enough" for most tasks. Those of us without fable-sized expense accounts noticed this quite a while back
SomeonesAccount•52m ago
Exactly! Composer 2/2.5 were amazing, cheap, and fast. Everyone else was Gaga about GPT 5.5 and such, while we were over here doing the work with less cost and more speed
jlkuester7•51m ago
Exactly. Even 32b parameter models you can run locally on consumer hardware are "good enough" at this point for some workflows!
jbjbjbjb•43m ago
I’ve been playing around with Luna, Terra and Sol and for the type of work I’ve been doing lately I actually think Sol is just a likely to trip up as Luna. Examples were Sol over assuming, persisting in the wrong direction, over engineering a little script to do some exploration of api. They can all be fixed but it’s a waste of tokens, I rather have Luna do it because course correction on small pieces of work is cheaper.
scoring1774•13m ago
I've found the distinction to be in how much I care about how the final product looks. If I want high-quality code I typically find a smaller model with a well-designed spec to do better, if I want it to just run and produce something close to my vague description typically Sol does better. For most actual business use-cases I think the first is likely better but the experimentation speed up with the frontier is very nice.
tosh•57m ago
I think we'll see more of this soon

replit is already leading the way with free luna usage

caust1c•46m ago
IMO big models are not a product in and of themselves. Inference is just a new type of compute. I'm confident that in two or three years, every product will have inference capabilities integrated into the experience, and models will become less and less distinctive from one another.

What most products need from a model is a pretty short list: the ability to make tool calls well, accurate recall, and the ability to follow directions without wavering (whether or not those directions are baked into the weights or provided in a system prompt). That covers 95% of inference utility in products.

We're nearly there, and I believe these capabilities will fit on small models.

Because of this though, I predict hardware demand will stay high despite demand for "hosted" inference dropping. Unless there's some regulatory shenanigans that step in to say otherwise.

NitpickLawyer•42m ago
> But I also think the demand for "fast/cheap/good-enough" models is just about to take off.

There's a sort of "revelation" I had in ~early '24 when I used a 7B local model with a library called Guidance (initially out of MS, then the team moved) to create a flow where the model would receive pseudocode for tests, first write the tests, and once I approved then started writing code until the tests passed. This was before "thinking" models, and yet using that library I was able to "guide" the model in the required "prompt / instruct" context such that it was working towards completion, and I saw the first things like we see now in the thinking traces "oh, test x doesn't pass because blah, I need to..." and so on.

Anyway, the revelation was "even if the models never improve, I'll have years of fun finding out all the ways I can use these things". And, obviously, the models improved a lot since then. But I think that revelation can still be applied, as a sort of "truism". We have, right now, access to things that 10-20 years ago would be considered magic. We are still finding ways of cobbling together systems with glue, duct tape and prayers and find new things they can do.

I think the "good-enough" stage has come not just for API models (cheap, fast, etc) but for local as well. Even if slower, even if clunkier, but they are good enough for a set of ever increasing tasks, and what's more it's incredibly fun to work with them.

LoveMistral•25m ago
Same. Mistral 7b has been more than I ever needed for text for years now.

Unless you must 1-shot with no harness it’s the same amount of power, maybe more because the big “good” models make too many assumptions and tend to become rigid.

Mistral 7b can do anything, and it’s basically instant even on an M3

frigidwalnut•18m ago
Sounds interesting. Can you give more details on your workflow and what tasks you use it for?
Almondsetat
agcat•41m ago
I like the analogy on ways to make small model useful.
zatkin•40m ago
Maybe I'm being super reductive here, but operating small models at the core of your business kind of moves the needle from making external API calls (against frontier models) to running internal API calls (against your locally-run models). It seems like if we want local models to take off, it will need to become easier to run local models for cheap. I'm thinking like reducing the barrier of entry for running "local models" in the cloud providers like DigitalOcean, AWS, etc.
malfist•37m ago
You should be glad to know digital ocean already offers this
NickNaraghi•39m ago
> Across his various startups, Peter has seen two kinds of work: > 1. the "IQ 180" work. some mad scientist genius type comes up with some crazy solution you've never thought of. > 2. the "token spewer" work. being ultra responsive, pushing the ball forward across dozens of different fronts.

Interesting comp to pg's Maker's Schedule, Manager's Schedule https://www.paulgraham.com/makersschedule.html

I'm curious about not only which of these roles models will fill, but also how they will empower us to be in the mode we prefer.

michael0church•37m ago
It makes sense that we’ll see “room at the bottom” strategies. Currently, large parameter counts seem to be slush funds of world knowledge, language skills (because language’s nuances and open vocabulary make it high-dimensional), and reasoning primitives, the general belief being that the latter takes up the least space in the model.

There are many applications where world knowledge is unnecessary or even a negative, and in which only a small amount of language skill is necessary, and there we can expect small models more intelligently used to beat large ones naively used.

LPisGood•36m ago
Small amounts of world knowledge seems like it would inherently be tied to more hallucinations.
TJTorola•7m ago
Perhaps we'll get to a point where believing any un-sourced information from an LLM will feel crazy. I don't want my model to know more than it needs to perform logic and use tools. Once it is capable of using tools I would much rather it looked up information or sourced it from existing context rather than just divine it from it's weights.
giraffe_lady•9m ago
Everyone wants this to be it but over and over we discover that the bigger a model is the better it is at all tasks, even ones far outside the domain it was optimized for. IE claude fable is better at writing both code and prose than smaller code- and prose-specific models.

The way vision and language models converge into the same geometric space should be extremely alarming for the "you don't need global knowledge for local tasks" type dreams.

And to be clear I'm not saying that smaller models don't or can't work well, or that we shouldn't be heading in this direction. And it's not quite the case that broad knowledge is strictly necessary. But it never seems to be negative! And so far it is the best way we've found to do... everything. Small models are good to the extent they are like big models, not to the extent that they are small.

wxw•35m ago
100% agreed. Small, cheap, and hosted models. Luna (and open weight models and others) is ridiculously cheap @ $0.2/$1.2, easily accessible, and more than good enough for basic use cases (e.g. summarization, simple tool calling, etc.).
hartator•33m ago
I have trouble seeing the points of using less capable models.

I just want the smartest, best, and most capable models. It feels smaller models for speed and cost are just transitions towards better hardware allowing the very best model.

trvz•27m ago
First, smaller models are fun for hackers: you can run them locally, or run them faster.

Second, when cloud models become unavailable or otherwise deteriorate, these will be all you have. May as well prepare.

polotics•26m ago
Can you define your use of the word 'smartest' here just in case some of us don't quite know what you mean?
shafyy•22m ago
Some reasons: - Smaller models will always be cheaper - Smaller models will always use less energy, therefore better for the environment

It's a bit like saying you always want the fastest and best car; Sure, you can have it if you keep paying for it. But a small car will also get you from A to B, will use less gas and will be much cheaper.

tartuffe78•8m ago
Cost is the point
jmtulloss•22m ago
I forked my Big Serious Harness™ that models construction projects into a harness for building a vibe coded family assistant. I couldn't figure out how to make the toy operate at toy prices until Luna. Now you can vibe code all the little apps you might want for your fam for like $5 and operate it day to day for a few cents.
throwaway63467•19m ago
I’m kind of cautiously excited for the next five to ten years, with these AI chips becoming incredibly fast and RAM capacities ramping up its in the cards that we’ll have chips like today’s ATMEL microprocessors that fit on a single board computer and can run small models locally, then all our gizmos can have local AI and I can have a truly intelligent home. Of course there will be a huge push to put all of it in the cloud but maybe we have a chance to take this technology home for good as it’s hard to imagine people will submit to this kind of surveillance required for AI home automation 24/7 (then again I might be wrong). Exciting times.
possibilistic•14m ago
> Peter runs multiple companies. Beyond Segment, he's raised $100m+ for Charm Industrial, and just recently closed a Series A for Revoy. He's incredibly organized and efficient with his time.

You can do this before an exit?

mattmaroon•10m ago
The demand for fast, cheap, good enough models has always been borderline infinite, it’s the supply that’s going to take off.
•
8m ago
What kind of work are you doing? For example, if I have some code in the hot path and I want to do all the usual tricks to help the compiler vectorize it, such a small model is not able to do much.
jermaustin1•16m ago
To me, most local models work just fine for anything you can be patient for. If I want something quicker, I will go to a SOTA model via API, but with multiple 3090s, I have never really needed a hosted model for a lot of my experiments.

For code, they are great, but for creativity for NPC controllers, they leave something to be desired, but work well enough for testing, so I don't burn tokens until I'm actually playing my games.

But nothing one-shots a prototype better than Fable 5. I can have a prototype built in 30 minutes, hooked up to my local LLMs and Claude Code is very good at testing the interactions and even tuning the prompts of the NPCs for better experiences.

__float•4m ago
"with multiple 3090s" is quite a bit of burying the lede for "most local models work just fine", don't you think?