[HN Gopher] Show HN: Apfel - The free AI already on your Mac
___________________________________________________________________
Show HN: Apfel - The free AI already on your Mac
Github: https://github.com/Arthur-Ficial/apfel
Author : franze
Score : 629 points
Date : 2026-04-03 09:15 UTC (13 hours ago)
(HTM) web link (apfel.franzai.com)
(TXT) w3m dump (apfel.franzai.com)
| skrun_dev wrote:
| Notes.app handles big notebooks without choking on storage?
| p1anecrazy wrote:
| Really like demo cli tools description. Are they limited by the
| context window as well? What's your experience with log file
| sizes?
| franze wrote:
| the 2 hard limits of Appel Intelligence Foundation Model and
| therefor apfel is the 4k token context window and the super
| hard guardrails (the model prefers to tell you nothing before
| it tells you something wrong ie ask it to describe a color)
|
| parsing logfiles line by line, sure
|
| parsing a whole logfile, well it must be tiny, logfile hardly
| ever are
| reaperducer wrote:
| _the model prefers to tell you nothing before it tells you
| something wrong_
|
| If all LLMs did this, people would trust them more.
| khalic wrote:
| AFM models are very impressive, but they're not made for
| conversation, so keep your expectations down in chat mode.
| elcritch wrote:
| Any know if these only installed on Tahoe? I'm running Sequoia
| still and get an error about model not found.
| HelloUsername wrote:
| > _Apple Silicon Mac, macOS 26 Tahoe or newer, Apple
| Intelligence enabled_
| jonpurdy wrote:
| Yes, the model ships with Tahoe, not previous versions.
|
| I too would love to try this for simple prompts but won't be
| updating past Sequoia for the foreseeable future.
| als0 wrote:
| Same. What a disaster Tahoe is.
| swiftcoder wrote:
| Anyone tried using this as a sub-agent for a more capable model
| like Claude/Codex?
| LatencyKills wrote:
| The combined (input/output) context window length is 4K. Claude
| would blow through that even when trying to read and summarize
| a small file.
| knocte wrote:
| With a small/minimalistic harness like Pi maybe it works
| well?
| franze wrote:
| project started with
|
| trying to run openclaw with it in ultra token saving mode, did
| totally not work.
|
| great for shell scripts though (my major use case now)
| khalic wrote:
| If you're looking into small models for tiny local tasks, you
| should try Qwen coder 0,5B. It's more of an experiment, but it
| can output decent functions given the right context
| instructions.
| xenophonf wrote:
| > _[Qwen coder 0,5B] can output decent functions given the
| right context instructions_
|
| Can you share a working example?
| khalic wrote:
| So... a prompt? I'm not on my laptop but I hooked it to
| cmp.nvim, gave it a short situational prompt, +- 10 lines,
| and started typing. Not anywhere near usable but with a
| little effort you can get something ok for repetitive
| tasks. Maybe something like spotting one specific code
| smell pattern. The advantage is the ridiculous T/s you get
| coredog64 wrote:
| I was thinking about the other way: Could you use this in front
| of Claude to summarize inputs and so reduce your token counts?
| gigatexal wrote:
| It's a very small model but I've been playing with it for some
| time now I'm impressed. Have we been sleeping on Apple's models?
|
| Imagine they baked Qwen 3.5 level stuff into the OS. Wow that'd
| be cool.
| thenthenthen wrote:
| The vision models and OCR are SUPER
| bombcar wrote:
| Apparently the Overcast guy build a beowulf cluster of Mac
| minis to use the Apple transcription service.
|
| https://www.linkedin.com/posts/nathangathright_marco-arment-...
| gigatexal wrote:
| Yeah heard him talk about that. 48 or so 16GB m4 Mac minis.
| Insane. The Beowulf lives
| xp84 wrote:
| The ATP episode where he talked about this was incredibly
| fascinating. Marco is such a role model to me - he has a
| complete immunity to fads and trends, and just does things
| the way he wants to. He adopts the 'new and cool' things only
| when they have a real benefit.
| gigatexal wrote:
| I largely subscribe to the use boring tech ethos ... but
| php? come on man.
|
| and yet... successful people have used it to build really
| successful things: Facebook, Tumblr (I think), the things
| Marco's been involved with.
|
| I just dunno outside of meta should we really be pushing
| php with all its flaws? or is it still flawed and I need to
| update my priors?
| bombcar wrote:
| There's something major to be said for going to war with
| the tools you have.
|
| And living with decisions made 15 years ago may be much
| more successful than trying to change horses mid-stream.
| FinnKuhn wrote:
| For small tasks this seems perfect. However it being limited to
| English from what I can tell is quite a downsite for me.
| trollbridge wrote:
| It can work in other languages? % apfel
| --model-info apfel v0.6.25 -- model info + model:
| apple-foundationmodel + on-device: true (always)
| + available: yes + context: 4096 tokens +
| languages: zh, en, nl, zh, es, es, ja, en, pt, da, fr, it,
| nb, vi, tr, en, de, fr, es, pt, ko, sv, zh + framework:
| FoundationModels (macOS 26+)
|
| Just use the language you want when prompting it, like other
| LLMs? % apfel "Gib mir ein Rezept fur
| Currywurst." Naturlich! Hier ist ein einfaches Rezept
| fur Currywurst: ### Zutaten: - **Fur die
| Wurste:** - 500 g Bratwurste (z. B. Frankfurter
| Wurste)
|
| (note: clipped most of the reply, since I assume most of us
| here don't actually need an LLM-generated recipe)
| FinnKuhn wrote:
| Ok, then it might just be pretty bad at German Sie/Du. It
| manages to translate texts, but is unable to change between
| Du and Sie. interesting limitation I haven't encountered
| anywhere before. Even DeepL was able to swap between them.
|
| Here is the chat I used to test this again to highlight
| this limitation:
|
| you> write a short email to john asking for a good time to
| meet ai> Subject: Let's Catch Up Soon!
|
| Hi John,
|
| I hope you're doing well! I'd love to grab a coffee or meet
| up for a chat sometime soon. When do you have some free
| time this week?
|
| Looking forward to hearing from you!
|
| Best, [Your Name]
|
| you> Please translate this to German ai> Betreff: Lass uns
| bald treffen!
|
| Hallo John,
|
| Ich hoffe, es geht dir gut! Ich wurde gerne Kaffee trinken
| oder uns fur ein Gesprach treffen. Wann haben Sie in dieser
| Woche freie Zeit?
|
| Ich freue mich darauf, von Ihnen zu horen!
|
| Liebe Grusse [Dein Name]
|
| you> Bitte mit Sie Form anstatt Du ai> Betreff: Lass uns
| bald treffen!
|
| Hallo John,
|
| ich hoffe, es geht dir gut! Ich wurde gerne Kaffee trinken
| oder uns fur ein Gesprach treffen. Wann haben Sie in dieser
| Woche freie Zeit?
|
| Ich freue mich darauf, von Ihnen zu horen!
|
| Liebe Grusse [Dein Name]
|
| you> Bitte alles mit Du anstatt Sie ai> Betreff: Lass uns
| bald treffen!
|
| Hallo John,
|
| ich hoffe, es geht dir gut! Ich wurde gerne Kaffee trinken
| oder uns fur ein Gesprach treffen. Wann haben Sie in dieser
| Woche freie Zeit?
|
| Ich freue mich darauf, von dir zu horen!
|
| Liebe Grusse [Dein Name]
|
| My config: % apfel v0.6.13 -- model info + model: apple-
| foundationmodel + on-device: true (always) + available: yes
| + context: 4096 tokens + languages: en, fr, zh, zh, ko, en,
| fr, de, tr, pt, it, pt, es, nl, da, nb, zh, es, es, ja, sv,
| vi, en + framework: FoundationModels (macOS 26+)
| ramon156 wrote:
| Cool tool but I don't get why these websites make idiotic claims
|
| > $0 cost
|
| No kidding.
|
| Why not just link the GH Github: https://github.com/Arthur-
| Ficial/apfel
| ffsm8 wrote:
| He did?
|
| https://news.ycombinator.com/item?id=47624647
| brians wrote:
| I've seen several projects like this that offer a network server
| with access to these Apple models. The danger is when they expose
| that, even on a loop port, to every other application on your
| system, including the browser. Random webpages are now shipping
| with JavaScript that will post to that port. Same-origin
| restrictions will stop data flow back to the webpage, but that
| doesn't stop them from issuing commands to make changes.
|
| Some such projects use CORS to allow read back as well. I haven't
| read Apfel's code yet, but I'm registering the experiment before
| performing it.
| stingraycharles wrote:
| I don't think many browsers will allow posting to 127.0.0.1
| from a random website. What's the threat model here?
| brians wrote:
| I think any browser will allow it but not allow data read
| back.
| mememememememo wrote:
| Isn't there a CORS preflight check for this? In most cases.
| I guess you could fashion an OG form to post form fields.
| But openai is probably a JSON body only.
|
| The default scenario should be secure. If the local site
| sends permissive CORS headers bets may be off. I would need
| to check but https->http may be a blocker too even in that
| case. Unless the attack site is http.
| airza wrote:
| There is a CORS preflight check for POST requests that
| don't use form-encoding. It would be somewhat surprising if
| these weren't using JSON (though it wouldn't be that
| surprising if they were parsing submitted JSON instead of
| actually checking the MIME-type which would probably be bad
| anwyay)
| btown wrote:
| FWIW this _was_ the status quo (webpage could ping
| arbitrary ports but not read data, even with CORS
| protections) - but it is changing.
|
| This is partially in response to
| https://localmess.github.io/ where Meta and Yandex pixel JS
| in websites would ping a localhost server run by their
| Android apps as a workaround to third-party cookie limits.
|
| Chrome 142 launched a permission dialog:
| https://developer.chrome.com/blog/local-network-access
|
| Edge 140 followed suit: https://support.microsoft.com/en-
| us/topic/control-a-website-...
|
| And Firefox is in progress as well, though I couldn't find
| a clear announcement about rollout status:
| https://fosdem.org/2026/schedule/event/QCSKWL-firefox-
| local-...
|
| So things are getting better! But there was a scarily long
| time where a rogue JS script could try to blindly poke at
| localhost servers with crafty payloads, hoping to find a
| common vulnerability and gain RCE or trigger exfiltration
| of data via other channels. I wouldn't be surprised if this
| had been used in the wild.
| layer8 wrote:
| Restricting such access it is still a work in progress:
| https://wicg.github.io/local-network-access/
| brians wrote:
| They offer it as an option but default it to false! This is
| still a --footgun option but it's the least unsafe version I've
| seen yet! Well done, Apfel authors.
| franze wrote:
| thx for the report - a totally valid attack vector i was not
| aware of before, should be fixed https://github.com/Arthur-
| Ficial/apfel/releases/tag/v0.6.23 - see also new
| https://github.com/Arthur-
| Ficial/apfel/blob/main/docs/server...
| robotswantdata wrote:
| Keep seeing similar mistakes with vibe coded AI & MCP projects.
| Even experienced engineers seem oblivious to this attack vector
| snarkyturtle wrote:
| Noting that there's an option to require a Bearer token to the
| API
| Oras wrote:
| I like the idea and the clarity to explain the usage, my question
| would be: what kind of tasks it would be useful for?
| khalic wrote:
| Making a sentence out of a json
| convexly wrote:
| I like the approach of running everything locally. I'm strongly
| of the opinion that the privacy angle for local models is going
| to keep getting stronger and more relevant. The amount of
| articles that come out about accidents happening because of
| people handing too much context to cloud models the more self
| reinforcing this will become.
| aswanson wrote:
| That's the way things have to go. Business risk is too high
| having everything ran over exposed networks.
| lukewarm707 wrote:
| what i say about this, is that an llm is just a big file,
| there is nothing 'not private' about it.
|
| if you are happy with off-prem then the llm is ok too, if you
| need on-prem this is when you will need local.
| zahlman wrote:
| > an llm is just a big file, there is nothing 'not private'
| about it.
|
| The private thing is the prompt.
|
| But also, a local LLM opens up the possibility of agentic
| workflows that don't have to touch the Internet.
| lukewarm707 wrote:
| local is best for privacy, but i personally think you don't
| need to go local.
|
| anthropic, google, openai etc, decided that their consumer ai
| plans would not be private. partly to collect training data,
| the other half to employ moderators to review user activity for
| safety.
|
| we trust that human moderators will not review and flag our
| icloud docs, onedrive or gmail, or aggregate such documents
| into training data for llms. it became the norm that an llm is
| somehow not private. it became a norm that you can't opt out of
| training, even on paid plans (see meta and google); or if you
| can opt out of training, you can't opt out of moderation.
|
| cloud models with a zero retention privacy policy are private
| enough for almost everyone, the subscriptions, google search,
| ai search engines are either 'buying' your digital life or
| covering themselves for legal reasons.
|
| you can and should have private cloud services, and if legal
| agreement is not enough, cryptographic attestation is already
| used in compute, with AWS nitro enclaves and other providers.
| inetknght wrote:
| > _i personally think you don 't need to go local._
|
| I personally think everyone should default to using local
| resources. Cloud resources should only be used for expansion
| and be relatively bursty rather than the default.
| mark_l_watson wrote:
| For about two years I experimented with writing local apps
| using local LLMs, but I often had to blend in a commercial
| web search API to make my little experiments useful.
| mark_l_watson wrote:
| I pay $13/month for Proton's Lumo+ private chat LLM that
| contains an excellent built-in web search tool. I use it for
| everything non-technical, even just simple searching for
| local businesses, etc.
|
| As an enthusiastic reader of books like Privacy is Power and
| Surveillance Capitalism, it feels good to have a private tool
| that is ready at hand.
| djl0 wrote:
| do you have any provider recommendations? I've experimented
| with this on runpod serverless, but I've been meaning to dig
| deeper before I feel comfortable with personal data.
|
| I saw a service named Phala, which claims to be actually no-
| knowledge to server side (I think). It was significantly more
| expensive, but interesting to see it's out there. My thought
| was escaping the data-collection-hungry consumer models was a
| big win.
| sebastiennight wrote:
| > anthropic, google, openai etc, decided that their consumer
| ai plans would not be private. partly to collect training
| data, the other half to employ moderators to review user
| activity for safety.
|
| That's two halves of "why", sure.
|
| Another interesting half would be that those companies have
| US military officers on their boards, and LLMs are the
| ultimate voluntary data collection platform, even better
| trojan horses than smartphones.
|
| Yet another "half" could be how much enterprise value might
| be found by datamining for a minute or two... may I suggest
| reading a couple of Martha Wells books.
| ge96 wrote:
| The other thing, is encrypted inferencing a thing/service
| currently? I want to run my own models locally just because if
| I'm going to be chatting to it about my day to day life why
| send it to a server in plaintext.
| lukewarm707 wrote:
| encrypted inferencing, meaning homomorphic encryption: no,
| it's not solved.
|
| cryptographic confirmation of zero knowledge: yes.
|
| the latter, based on trust in the hardware manufacturer and
| their root ca. so, encrypted if you trust intel/nvidia to
| sign it.
|
| there are a few services, phala, tinfoil, near ai, redpill is
| an aggregator of those
| cousin_it wrote:
| It's only half of the solution though. If the models are
| trained in a closed way, they can prioritize values encoded
| during training even if that's not what you want (example: ask
| the open Chinese models about Tiananmen). It's not beyond
| imagining that these models would e.g. try to send your data to
| authorities or advertisers when their training says so, even if
| you run them locally.
|
| So the full solution would be models trained in an open
| verifiable way _and_ running locally.
| wrxd wrote:
| The model is only generating tokens without touching the
| network at all, right? How would it send data away?
| procaryote wrote:
| Theoretically, by taking the opportunity to inject an
| exfiltration mechanism if you ask it to write code for you
| kg wrote:
| Lots of people I know run models in "yolo" mode or the
| equivalent as well, which means it could just invoke curl
| or telnet to exfiltrate data.
| hombre_fatal wrote:
| Another angle is when you're passing untrusted content to the
| AI service, e.g. anything from using it to crawl websites to
| spam-detection on new forum user posts.
|
| You can trigger the the service's ToS violation or worse, get
| tipped off to law enforcement for something you didn't even
| write.
| Xenoamorphous wrote:
| > I like the approach of running everything locally. I'm
| strongly of the opinion that the privacy angle for local models
| is going to keep getting stronger and more relevant.
|
| In HN circles perhaps. Average Joes don't care.
| arendtio wrote:
| For those who don't know, 'Apfel' is the German word for Apple.
| gherkinnn wrote:
| And for those who did know that and want to know more, the
| shift from apple - apfel and water -> wasser happened during
| the High German consonant shift.
|
| https://en.wikipedia.org/wiki/High_German_consonant_shift
| VanTodi wrote:
| Just a small thing about the website: your examples shift all the
| elements below it on mobile when changing, making it jump
| randomly when trying to read.
| gherkinnn wrote:
| Now this is a development I like.
|
| With the Claude bug, or so it is known, burning through tokens at
| record speed, I gave alternative models a try and they're mostly
| ... interchangeable. I don't know how easy switching and low
| brand loyalty and fast markets will play out. I hope that local
| LLMs will become very viable very soon.
| naravara wrote:
| Yeah I don't think the models are meaningfully differentiated
| outside of very specific edge cases. I suspect this was the
| thinking behind OpenAI and Facebook and all trying to lean hard
| into presenting their chatbots as friends and romantic
| partners. If they can't maintain a technical moat they can try
| to cultivate an emotional one.
| g-mork wrote:
| Saw a comment here yesterday referencing the Attention Is All
| You Need paper title in a tongue in cheek way. Kinda fun to
| imagine the friend/romance angle is just a bunch of socially
| awkward folk at OpenAI misinterpreting the original paper
| acuozzo wrote:
| > very specific edge cases
|
| Mathematics is hardly an edge case, but SOTA models differ
| wildly in their ability to write proofs for unsolved
| problems.
|
| Models also differ wildly in tasks like decompilation for
| reverse engineering.
|
| Also, so far, the only model I've found which can competently
| write PTX for SM100 CUDA devices is GPT-5.4pro, but I'm
| willing to admit that this is more of an edge case than the
| aforementioned.
|
| AFAICT, the extent to which someone finds models
| interchangeable is inversely proportional to the novelty of
| their work.
| m-s-y wrote:
| A serious project would do the work to be delivered via the
| native homebrew repository, not a "selfhosted" one.
| brtkwr wrote:
| Isn't the whole idea of "home brew" to enable hackers and
| enthusiasts to easily share what they built?
| post-it wrote:
| Is this you signing up as a packager or
| nose-wuzzy-pad wrote:
| Does the local LLM have access to personal information from the
| Apple account associated with the logged-in user? Maybe through a
| RAG pipeline or similar? Just curious if there are any risks
| associated with exposing this in a way that could be exploited
| via CORS or through another rogue app querying it locally.
| franze wrote:
| no. the on device foundationmodels framework that apfel uses
| does not have access to personal information from the apple
| account. the model is a bare language model with no built in
| personal data access.
|
| apple does have an on device rag pipeline called the semantic
| index that feeds personal data like contacts emails calendar
| and photos into the model context but this is only available to
| apples own first party features like siri and system summaries.
|
| it is not exposed through the foundationmodels api.
| phplovesong wrote:
| This is pretty cool. My bet is that we have more LLMs running
| locally when its possible, either thru "better hardware as
| default" or some new tech that can run the models on commodity
| hardware (like apple silicon / equivalent PC setup).
| alwinaugustin wrote:
| Read Austria as Australia and thought this as an April fool
| nottorp wrote:
| > Starting with macOS 26 (Tahoe), every Apple Silicon Mac
| includes a language model as part of Apple Intelligence.
|
| So you have to put up with the low contrast buggy UI to use that.
| mattkevan wrote:
| As an experiment I built a prototype chatbot app that uses the
| built-in LLM. It's got a small context window, but is
| surprisingly capable and has tool-calling support. Without too
| much effort I was able to get it to fetch weather data, fetch and
| summarise emails, read and write reminders and calendar events.
| joriskok1 wrote:
| How much storage does it take up?
| franze wrote:
| 4mb download, after install about 15mb, model is already on
| your mac with mac os x tahoe
| Barbing wrote:
| Just discovered iOS shortcuts has a native action called "use
| model" that lets you use local, Apple cloud, or ChatGPT-- before
| that I would have agreed with the author about being locked
| behind Siri (natively)
| api wrote:
| BoltAI also does this, but a CLI tool is nice.
|
| It's a nice LLM because it seems fairly decent and it loads
| instantly and uses the CPU neural engine. The GPU is faster but
| when I run bigger LLMs on the GPU the normally very cool M series
| Mac becomes a lap roaster.
|
| It's a small LLM though. Seems decent but it's also been safety
| trained to a somewhat comical degree. It will balk over safety at
| requests that are in fact quite banal.
| pbronez wrote:
| Digging into this, found Apple's release notes for the Foundation
| Model Service
|
| https://developer.apple.com/documentation/Updates/Foundation...
|
| They released an official python SDK in March 2026:
|
| https://github.com/apple/python-apple-fm-sdk
| als0 wrote:
| Is this for Tahoe only? I'm still clutching onto Sequoia
| linsomniac wrote:
| Yes, it says on that page that it uses Apple Intelligence from
| Tahoe. I'm also hanging onto Sequoia, though I'm ready to make
| the leap any time here.
| crena wrote:
| MacBook Neo forced me to finally make the jump, and it turns
| out that I, much like the engineers at Apple, don't really
| care about the spit and finish anymore. Third-party
| applications handle everything else. Also, I was happy to
| find that Divvy still runs just fine under Rosetta.
| anentropic wrote:
| Yeah seems to need Tahoe (I'm on Sequoia):
| dyld[71398]: Library not loaded: /System/Library/Frameworks/Fou
| ndationModels.framework/Versions/A/FoundationModels
| Referenced from: <32818E2F-CB45-3506-A35B-AAF8BDDFFFCE>
| /opt/homebrew/Cellar/apfel/0.6.25/bin/apfel (built for macOS
| 26.0 which is newer than running OS) Reason: tried: '
| /System/Library/Frameworks/FoundationModels.framework/Versions/
| A/FoundationModels' (no such file), '/System/Volumes/Preboot/Cr
| yptexes/OS/System/Library/Frameworks/FoundationModels.framework
| /Versions/A/FoundationModels' (no such file), '/System/Library/
| Frameworks/FoundationModels.framework/Versions/A/FoundationMode
| ls' (no such file, not in dyld cache)
| rbbydotdev wrote:
| Would really love to see a web api standard for on device llms.
| This could get us closer. Some in-browser language model usage
| could be very powerful. In the interim maybe a little protocol
| spec + a discovery protocol used with browser plugins, web apps
| could detect and interface with on-device llms making it
| universally available.
| dchest wrote:
| https://webmachinelearning.github.io/prompt-api/
|
| Already in Chrome as an origin trial:
| https://developer.chrome.com/docs/ai/prompt-api
| walthamstow wrote:
| You have to enable Apple Intelligence so that's a hard no from
| me. I'll stick to LM Studio and gpt-oss/qwen. Very cool project
| though.
| btucker wrote:
| I hacked this together last fall to let you use Apple Foundation
| Models with llm: https://github.com/btucker/llm-apple . To enable
| that I built python bindings with Claude Code:
| https://github.com/btucker/apple-foundation-models-py
|
| Unfortunately, I found the small context window makes the utility
| pretty limited.
| troyvit wrote:
| Yeah I think you hit on the head a good way to use it though.
| I'm not on MacOS but KDE has a little tool called krunner[1]
| that lets you perform simple tasks from a small pop-up on your
| desktop. It would be cool if I could do slightly agentic things
| from there with a local model like ask what the capital of
| Austria is, or what's the current exchange rate between two
| currencies.
|
| Then save the heavy lifting for the big boys.
|
| [1] https://userbase.kde.org/Plasma/Krunner
| jasomill wrote:
| I've used Alfred[1] for many years on macOS, and more
| recently the PowerToys Command Palette[2] on Windows.
|
| Both are easily extensible, so you could certainly wire
| either up to the local (or hosted) LLM of your choice.
|
| Side note: I just noticed the extension example in the
| Command Palette demo reel searches HN.
|
| [1] https://www.alfredapp.com/
|
| [2] https://learn.microsoft.com/en-
| us/windows/powertoys/command-...
| mark_l_watson wrote:
| I have been using Apple's built-in system LLM model for the last
| 7 or 8 months. I like the feature that if it needs to, it
| occasionally uses a more powerful secure private cloud model. I
| also write my own app to wrap it.
| donmb wrote:
| Local AIs are the future in times of limited resources. This
| could be the beginning of something big. I like that Apple opens
| up like this. Hopefully more to come.
| enjoyitasus wrote:
| completely agree.
| Phemist wrote:
| Nice! The example should imo say
|
| apfel -o json "Translate to German: apple" | jq .content
| franze wrote:
| Thx. Good point. Done.
| Multiplayer wrote:
| Started using this earlier this week. I built a backtesting
| benchmark tool to compare a mix of frontier and open-source
| models on a fairly heavy data analysis workflow I'd been running
| in the cloud.
|
| The task is basically predicting pricing and costs.
|
| Apple's model came out on top--best accuracy in 6 out of 10 cases
| in the backtest. That surprised me.
|
| It also looks like it might be fast enough to take over the whole
| job. If I ran this on Sonnet, we're talking thousands per month.
| With DeepSeek, it's more like hundreds.
|
| So far, the other local models I've tried on my 64GB M4 Max
| Studio haven't been viable - either far too slow or not accurate
| enough. That said, I haven't tested a huge range yet.
| sys_64738 wrote:
| Tahoe+ only
| EddieLomax wrote:
| This is similar to something I was playing around with last
| month-- basically just a CLI for accessing the foundational
| models.
|
| https://github.com/ehamiter/afm
|
| It's really handy for quick things like "what's the capital of
| country x" but for coding, I feel that it is severely limited.
| With such a small context it's (currently) not great for
| complicated things.
| divan wrote:
| What's the easiest way to use it with on-device voice model for
| voice chat?
| windsurfer wrote:
| https://github.com/Arthur-Ficial/apfel-gui uses on-device
| speech-to-text and text-to-speech
| divan wrote:
| Thanks, tried it, but it's crashes on clicking the microphone
| icon. Default `make install` for some reason tries to install
| it to /usr, I changed that and after torturing more mature
| coding LLMs for 20 minutes, made it running with mic/sound.
|
| The mic button requires clicking to transcribe and start
| listening again, and default voice is low-quality (I assume
| it can be configured).
|
| In general I'm looking for a way to try the on-device hands-
| free voice mode.
| contingencies wrote:
| https://handy.computer
| satvikpendem wrote:
| How does this model compare against other local models like Qwen
| run through LMStudio?
| frontsideair wrote:
| > Apple locked it behind Siri. apfel sets it free
|
| This doesn't feel truthful, it sounds like this tool is a hack
| that unlocks something. If I understand it correctly, it's using
| the same FoundationModels framework that powers Apple
| Intelligence, but for CLI and OpenAI compatible REST endpoint.
| Which is fine, just the marketing goes hard a bit.
|
| > Runs on Neural Engine
|
| Also unsure if this runs on ANE, when I tried Apple Intelligence
| I saw that it ran on the GPU (Metal).
| reaperducer wrote:
| apfel "How many shopping days until Christmas?"
|
| _To determine how many shopping days until Christmas, you 'll
| need to know the current date and the date of Christmas.
| Christmas is on December 25th, so you can subtract the current
| date from this date to find the number of shopping days. If you
| provide the current date, I can help you calculate that._
| apfel "How many shopping days until Christmas? Today is April
| 3rd."
|
| _To calculate the number of shopping days until Christmas, we
| first need to determine the date of Christmas this year.
|
| Christmas in 2023 is on December 25th.
|
| Starting from April 3rd, we calculate the days:
|
| 1. April: 30 days (from April 3rd to April 30th) 2. May: 31 days
| 3. June: 30 days 4. July: 31 days 5. August: 31 days 6.
| September: 30 days 7. October: 31 days 8. November: 30 days
|
| Adding these up gives us:
|
| 30 + 31 + 30 + 31 + 31 + 30 + 31 + 30 = 252 days
|
| Christmas is on December 25th, so we subtract 252 days from
| today, April 3rd:
|
| 252 - 252 = 0 days
|
| So, there are 0 shopping days left until Christmas, but since
| Christmas is already past, there are no shopping days left._
|
| Yep, feels like Siri.
| xp84 wrote:
| Real experience I've had:
|
| "Text Carol bring me a glass of water please"
|
| "I'm sorry, I don't see a 'Carol Bring' in your contacts"
| 6persimmon wrote:
| Almost about to try it until I saw this. If it's Siri the Silly
| don't even make up for the opportunity cost.
| CharlesW wrote:
| This isn't a good test for any model since LLMs can't math
| (even though frontier models can sometimes correctly simulate
| mathing), which is why one would always use a tool for this.
| furyofantares wrote:
| Looks like a nice wrapper around the APIs. Extremely oversold
| landing page, very marketing heavy for what it is. You can
| actually make nice looking landing pages that are about 10% the
| size of this and more straightforward, rather than some mimicry
| of a SaaS that's trying desperately to sell you something. Makes
| it easier for you to review the content for factuality too, and
| heck you couldn't even take ownership of some of the voice.
|
| Hard to know what to do with this. I'm interested in the project
| and know others who would be, but I feel like shit after being
| slopped on by a landing page and I don't wish to slop on my
| friends by sharing it with them. I suppose the github link is
| indeed significantly better, I'll share that.
| xandrius wrote:
| It's absolutely free and open source, no need to bash it like
| this.
| furyofantares wrote:
| I certainly don't feel entitled to anybody's effort, and I
| appreciate the project itself!
|
| It's just these flashy LLM-generated webpages are really
| misleading.
|
| It's filled with attention-grabbing LLM-filler. It presents
| itself as high-effort content, but in reality is unclear if
| it's accurate, unclear how much the author even put into
| READING the content they're asking me to read, and overall
| just really unclear what parts of the page are important to
| the author and what are just stuff the LLM inserted into a
| marketing card to make it look like a SaaS sales pitch.
| wrxd wrote:
| If you dislike the slop on the website so much you might also
| not like GitHub that much. It shows this software has been
| heavily co-written by Claude code
| furyofantares wrote:
| I'm aware of that! That's fine with me.
| lewisjoe wrote:
| Tempted to write a grammarly-like underline engine that flags
| writing mistakes across all apps and browser. Fully private
| grammarly alternative without even bundling an LLM!
| malshe wrote:
| That's a great idea. I would be very interested in using it of
| someone builds it.
| gurjeet wrote:
| Thank you for making it open source!
|
| Submitted a PR to prevent its installation on macos versions
| older than Tahoe(26), since I was able to install it on my older
| macos 15, but it aborted on execution.
|
| https://github.com/Arthur-Ficial/homebrew-tap/pull/1
| ronb1964 wrote:
| I'm a Linux user who wanted exactly this but for Linux -- so I
| ended up building it myself. It's called TalkType, it runs
| Whisper locally for offline speech-to-text. The privacy angle was
| a big reason I went local from the start -- I didn't want my
| voice being sent to anyone's server. Nice to see the same idea
| getting traction on Mac.
| yalogin wrote:
| This is great. A few questions come to mind, I need to go look
| up. Is the model an OpenAI one or home grown for Apple. And can I
| still use it if Siri is disabled?
| contingencies wrote:
| On a similar bent, I recently discovered Handy (cross-platform)
| which is very well implemented local voice input:
| https://handy.computer/ ... serious finger saver and ideal for
| LLM conversations
| rgbrgb wrote:
| love the simple website and typography. AI design or you?
| tasteful and fast animations. nice work and thanks for sharing!
| deadfox wrote:
| This is cool!
| karimf wrote:
| The big question is whether Apple can keep shipping new models
| constantly.
|
| AFAIK the current model is on par with with Qwen-3-4B, which is
| from a year ago [0]. There's a big leap going from last year
| Qwen-3-4B to Qwen-3.5-4B or to Gemma 4.
|
| Apple model is nice since you don't need to download anything
| else, but I'd rather use the latest model than to use a model
| from a year ago.
|
| https://machinelearning.apple.com/research/apple-foundation-...
| dangus wrote:
| I'm not sure why that's a question, it's just a downloaded
| file. You can even watch it download separately when you enable
| Apple Intelligence (it's not tied to OS updates from what I can
| tell).
|
| Of course I imagine Apple is not going to be the fastest mover
| in this regard. I'm not even sure they believe the product will
| be widely impactful anymore and may keep it relegated to a
| small list of popular use cases like photo touch ups and quick
| questions to Siri. For me the most useful parts of Apple's AI
| don't even require me to enable Apple Intelligence.
| xp84 wrote:
| > Referenced from: <32818E2F-CB45-3506-A35B-AAF8BDDFFFCE>
| /opt/homebrew/Cellar/apfel/0.6.25/bin/apfel (built for macOS 26.0
| which is newer than running OS)
|
| This actually looks really neat. I'll have to bookmark this for
| whenever I'm dragged kicking and screaming into the abomination
| that is "Tahoe."
| witnessme wrote:
| Interesting. How does this foundational model compares with other
| LLMs?
| contingencies wrote:
| 1. Hugely non-deterministic: repeat queries give vastly different
| responses. 2. Often returns incorrect and inconsistent results
| even for mathematical queries. 3. Often the responses include
| unwanted highlighting or presentation markup. 4. Defaults to
| German decimal notation.
| andy_xor_andrew wrote:
| I find the branding to be a little odd. Like, it should be a
| github page with a README that says "here's how to use this."
| Like, the full explanation of this project is right there in the
| HN title: "The free AI already on your Mac."
|
| I guess LLMs have made it too simple to instantly build startup
| landing page slop, which causes this? Like, do we need to see the
| github star count chart? Do we need all the buzzwords and stuff?
| You'd think this was a startup trying to get a billion dollar
| evaluation. It feels disingenuous.
|
| Maybe I'm just being a hater.
| mrbonner wrote:
| This is great. I think Apple bought Kuzu, a in memory graph
| database in late 2025 to support RAG in combine with their FM
| like this. Even with such as small model, a comprehensive context
| of our personal data in graph RAG would be sufficient for a PA
| system. Do we know if we can have access to this RAG data?
| newman314 wrote:
| Saw this in an another thread previously and immediately
| installed it.
|
| I have a new prompt to test LLMs much like simonw's pelican test.
|
| "What is 9:30am Taiwan time in US, Pacific?" For some reason, the
| answers are quite inconsistent but all wrong.
| ./apfel "what is 9:30am Taiwan time in US, Pacific?" Taiwan
| is 12 hours ahead of the Pacific Time Zone. Therefore, 9:30 AM
| Taiwan time would be 9:30 PM Pacific Time. Taiwan is 13
| hours ahead of the Pacific Time Zone. Therefore, 9:30 AM in
| Taiwan is 10:30 PM in the Pacific Time Zone. Taiwan is in
| the China Standard Time (CST) zone, which is 12 hours ahead of
| the Pacific Standard Time (PST) zone. Therefore, 9:30 AM in
| Taiwan is 9:30 PM in the Pacific. Taiwan is typically 11
| hours ahead of the Pacific Time Zone. Therefore, 9:30 AM in
| Taiwan is 8:30 PM in the Pacific Time Zone. Taiwan is 13
| hours ahead of the Pacific Time Zone. Therefore, 9:30 AM in
| Taiwan is 10:30 PM the previous day in the Pacific Time Zone.
| zopf wrote:
| Wow - the model really hallucinates without hesitation. I asked a
| number of "What do you know about [person, company, etc]?"
| questions and rather than realizing that it didn't know about
| them, it just made up an answer for every one of them.
| millionclicks wrote:
| Awesome idea. You should launch this on Buildfeed.co.
| lispisok wrote:
| I want access to the LLM xcode uses for swift auto completions
___________________________________________________________________
(page generated 2026-04-03 23:01 UTC)