Join the discussion
Write your take first — we'll ask for email only when you're ready to publish.
- Hacker News
- A dream of mine is to be able to host a LLM-powered video game that I can host on a home server running a decent mid-range GPU like the RTX 5060, and the LLM is fast and intelligent enough to make for a fun game experience for a few dozen concurrent players. People can ask for features and they just get made and added to the game on the fly for the lobby to enjoy. The hosting costs would be manageable enough that I don't have to charge anything for the game.
I think with one more year or so of small model progress, that might just be possible to accomplish.
by 2001zhaozhao - I’m doing this now. Homebrew DnD engine and mobile app chat interface. Rich 3d sim and haptics for the dice rolls. But otherwise no fancy graphics rendering. The juice is in the imaginative generate group story telling. Local real time STT on self hosted video chat, to convert the banter and side talk into next player action dice rolls. Qwen on an m3 ultra 96gb handles NPC, dm, bot party members, arc, consistency, history, memory, adventure beats. There’s even a mechanic for bards to compose an end-of-adventure ballad that compresses the shenanigans into verse.
Bootstrapped it with fable, opus or sol 5.6 when I had extra usage to burn over the last month. Playtesting it solo as both DM and PC was fun too. Character consistency and originality is remarkably compelling.
Upshot is that for the gaming hours that my friends have available, the social interpersonal gaming experience competes well for their AAA gaming time. There’s even some good async passive campaign continuation patterns that run in our signal thread to keep things going between sync sessions.
And players have requested features directly in the session transcript that have been direct implemented.
by U4E4 - Small is relative. I'm looking for models that I can with run around 100 MiB mark (RAM just for the weights) to demo what you can do with this little memory.
I know of SmolLM 2 which in Q4 is borderline regarding the size and rather dated. There is also TinyStories, which is also old and also focussed on children's stories.
Is there anything newer in this category? Or should I try to distill something down to this size?
by weinzierl - Luna max is suitable for like 90% of the kinds of code changes I want to make. I only find myself actually reaching for a Sol or Fable tier model if the problem is very complex. If you're willing to build the guardrails and do some extra planning, Luna is very capable.by dgunay
- Is there a resource somewhere that tells me, "Given that you have X Gb RAM on your laptop, these are the current local models you should consider trying and here's how to configure them to leave enough memory for other applications on your machine."?by Myrmornis
- I just asked my current LLM for that advice. Funny that they dont block it, I guess they are not very threatened.
For my anemic 6GB built-on 14GB Qwen seems to be the best bet, not great reviews but from my limited testing its pretty impressive.
by sandos - Several exist actually. Try whichllm.app or fitmyllm.com.by blueferret
- Are you familiar with Ollama [1]? It is a particularly easy to use tool to download and run local models. They sort models recent popularity and specify size for the various quantization levels.
I would try using ~1/2 your available ram and iterate from there.
If you have 32GB of RAM, I would give Qwen 3.8 a try. All you would have to do is run "ollama pull qwen3.8:27b" then "ollama run qwen3.8:27b". If you have 16GB of RAM, I would try Gemma 4.
[1] ollama.com
by bhelkey - Not that I know of, but https://www.canirun.ai/ might be of useby chrishare
- > 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?
by glimshe - We've been testing and developing a few openai based applications. Relatively simple stuff. I topped up my api credit with 10$ a few months ago. I just checked, we burned something like 61 cents over this period. About 126 requests over the last month. That's just us testing but it shows me that this stuff is relatively cheap. The reason we don't have a lot of requests is that most of this stuff pretty much works on the first try. This is not hard to build.
I think AI is an underused capability in applications. People come up with all these big plans to build agents. But even a simple button to "calculate a color palette that takes into account these symbolic names the user entered" is a simple prompt that comes back quickly and doesn't require a lot of reasoning. We use that with Luna and we give it some more context.
Image based stuff is more expensive but still quite reasonable. We do some things with Sol that work on images. It's not fast but the result is worth waiting for. These are features that save the user time and feel a bit magical. They just click a button, wait 20-30 seconds and then magic happens. Beats them clicking around for half an hour.
Local model hosting would force me into higher cost and overhead for less intelligence. That's only a good tradeoff if you are really going to use lots of tokens or you need this for other reasons (e.g. data security). But otherwise you just end up paying for underutilized GPU capacity.
- I think in general these smaller applications of AI are seriously overlooked even though they will actually add lots of economic value for companies and lots of quality-of-life features for users.by magimas
- A friend of mine told me earlier today that they had a discussion at work (a coding shop) about "downgrading" to luna from sol for cost reasons and that many were quite unhappy about this because they didn't want inferior tech to be forced upon them. Do they have a point? Is sol actually worth the extra cost? Especially if you ramp up the effort level?by teiferer
- Yes. If they don't like the cost then they should fire the "leader" who introduced the AI there to begin with.by dude250711
- It's silly to discuss it. Just do the evals.by brikym
- Is sol better?
Yes. Categorically. Anyone who tells you otherwise and that luna is “just as good” does not know what they are talking about.
Going from sol to luna is a downgrade.
It is not a question, it is a fact.
> Is sol actually worth the extra cost?
Is a question only you can answer, because it has no generic answer.
Right now, for me, being able to use sol is worth the cost, but using it all the time is not.
I’m sure going from using it to using luna feels rubbish; but there are realities about costs you have to face sooner or later.
Maybe like… give your team credits and make them pick the right tool for the job; and if they burn their credits on sol in 20 minutes, well, tough luck buddy, looks like you're coding by hand for the rest of the month.
Team will quickly shift. People hate losing access to ai.
- I choose to use Luna for most tasks because it is cost efficient, even though I get a pretty generous budget from my company.
Sometimes I will use Fable or Sol for large features/projects, or research/exploration.
I would not be at all happy if I were forced to use Luna, though. I’d probably start looking to leave. I don’t want to work somewhere where I don’t have choice over my tools.
- Luna as a doer, with a smarter model planning, can be a good compromise. Using sol for everything can be expensive without much gain, as a lot of steps don't need that sort of intelligence.by usef-
- I’m building an internal tool for our company, basically an agent to help with on-call and alerts via Slack. I have evals running across a few scenarios, and my favorite models so far are Sol medium and Luna xhigh.
Sol medium has been a nice balance between intelligence and response time. Luna xhigh can achieve similar scores on the evals, but it takes noticeably longer. My impression is that the higher reasoning effort helps compensate for the lower base intelligence.
Cost is definitely a big factor, but latency and intelligence matter too. If I had the budget, I’d take Sol medium over Luna xhigh.
From using both on real scenarios, Sol is noticeably better at navigating around issues, exploring alternatives, and being creative when the obvious approach doesn’t work. That matters quite a bit when you’re investigating live alerts, where the path to the root cause isn’t always straightforward.
by azuanrb - It is a good question. Luna is definitely a very capable model. Much more capable than the top SOTA models from 12 months ago. It definitely isn't at the same level as Sol, but you get 20x the tokens for the cost, and it has a much faster tokens/second rate.
If this is a cost conscious company where I'm going to get a fairly limited amount of Sol, or a nearly unlimited amount of Luna, I'm probably choosing Luna.
by pseudosavant - I don’t love the “forced upon them” framing; if that’s really how people are thinking about it then maybe they should pause and reflect for a moment yhat it isn’t their money being spent. Amd the default isn’t always having the latest and greatest, it’s not paying for anything at all.
Now, if the debate is really about which option is more cost effective, then we could easily run an A/B test to find out. Though TBH my instinct is that that experiment is likely to cost more than the potential cost savings.
What I will say is that my own sense from experimenting around in a non-rigorous way is that the answer depends on how you use the tool. For actual vibecoding you should always go for the SOTA model because it will need less oversight. It’s also less likely to get stuck in a vicious loop that fruitlessly wastes tokens. But for a more hands-on approach where you move in small, carefully planned increments that you review and test in human-comprehensible chunks, smaller models may be preferable. SOTA ones don’t do that much better when working that way, and the slower inference adds a detrimental amount of friction to the work cycle.
by bunderbunder - 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.
- Can we stick one of these in something that looks and sounds like HAL 9000 ?by mathgenius
- Same. I was in the doom AI crowd ~2 years ago, then I realized how many problems could be mitigated or solved by having these local models, and I became part of the cautiously excited crowdby tuesdaynight
- Is a "truly intelligent home" something I should want? None of the current generation of "smart" addons are true value adds - their entire purpose is data collection. The pretext is always absurdly thin. I just thought "I bet there's a wifi enabled microwave", googled it, and indeed, Samsung have released such a thing - you can control it with your voice! Wow! Never mind that you can't un/load food with your voice, or that you're never more than 10 feet from it in the kitchen anyway. Oh but there's an app. Of course there's an app. It "suggests recipes". Right.
I think hoping for a locally hosted "smart" home is backwards. They have no other function than to invade your space. "Smart" objects are agents, and they don't work for you.
by dTal - 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 backby swiftcoder
- > for most tasks
The word “most” is doing a lot of work here. On a percentage basis perhaps most tasks a typical SWE needs to do when they aren’t in meetings or writing docs are just glorified autocomplete. But that’s boring and that’s why people don’t usually talk about it.
People are addicted to chasing frontier models because they all have memories of spending a week on a deeply challenging algorithm problem or even have crazy complicated algorithms they cannot implement themselves and want to have the models achieve this technical breakthrough. It’s the kind of productivity boost from spending one week on a problem to spending one hour. In contrast the productivity boost from spending ten minutes to spending one minute just doesn’t occupy people’s mind.
by kccqzy - no they are not good enough for "most" tasksby dominotw
- Exactly. Even 32b parameter models you can run locally on consumer hardware are "good enough" at this point for some workflows!by jlkuester7
- I received access to AI in May of this year and used smaller models a lot in May, June, and July. Haiku 4.5 is terrible for coding. This month Luna was added and it's incredibly capable. I have to reach for a better model 1/50 times instead of 1/2 or 1/3 with Haiku. Luna is better and cheaper than the Haiku and Sonnet models I've been spending most of my time with.
It's safe to say I didn't come to this conclusion because I was "addicted to chasing frontier models"
by samspot