{"generated_at":"2026-11-12T18:49:20.246Z","source":{"name":"Hacker News public API","url":"https://github.com/HackerNews/API","attribution":"Stories and comments are public Hacker News submissions by their respective authors. HackerHub is an independent reader, not affiliated with Y Combinator.","content_is_untrusted":true},"story_id":50446520,"hn_url":"https://news.ycombinator.com/item?id=50446520","total_comments":550,"returned":20,"scope":"Up to the requested number of top-level comments in Hacker News ranked order. Deleted comments are omitted. Nested replies are not fetched; this is not the complete thread.","comments":[{"id":50447085,"author":"pen2l","created_at":"2026-11-10T22:18:53.000Z","text":"Apple, to me, is the winner in the end of this AI chapter.\nDistillation works: we can get models leaner, have better performance with lesser and lesser parameters, more efficient, running locally. Apple gets this and we'll see the result with M7 Max/ultra chips and I reckon this will be one of the titular apps when M7 line of chips are introduced.\nI guess it explains the haste Anthropic and OpenAI are in to IPO, cash out and leave the the retail investor holding the bag.","parent_id":50446520,"replies_count":38,"hn_url":"https://news.ycombinator.com/item?id=50447085"},{"id":50447481,"author":"simonw","created_at":"2026-11-10T22:57:19.000Z","text":"\"Harbor is currently in a private preview under Apple’s Claris brand\"\nThere's a name I haven't heard in a long time! Claris (always an Apple subsidiary, since being founded in 1987) renamed themselves to FileMaker in 1998, but apparently they rebranded back to Claris in 2019.\nLooks like they've mostly just released annual upgrades to FileMaker for the past decade, but they did add MCP to it last year: https://www.claris.com/blog/2025/why-mcp-matters-for-claris-...","parent_id":50446520,"replies_count":4,"hn_url":"https://news.ycombinator.com/item?id=50447481"},{"id":50450531,"author":"jph","created_at":"2026-11-11T07:10:24.000Z","text":"If you ask who would pay for Apple local models, I would. A Mac Studio high end 512GB costs ~$1000/month for 3 years versus my current run rate of ~$1000/month for five devs for Claude Ultra (or ~$10k/week for Claude Sonnet if I paid API rates rather than Ultra).\nIf/when Anthropic changes to pure API pricing, or for API-heavy projects, that's when I'd switch to more local models. Of course there are tradeoffs for speed, quality, upgradeability, etc.\nI currently run a Mac with 128GB and qwen3.8:27b-mlx. It hits the heat limit of the Mac when working on a 200k LOC codebase for security auditing and hardening for many hours at a time. This isn't as strong/fast as Sonnet, to be sure, yet the cost is only $6K one-time plus the usual electricity, maintenance time, etc.","parent_id":50446520,"replies_count":4,"hn_url":"https://news.ycombinator.com/item?id=50450531"},{"id":50451240,"author":"thibsrn","created_at":"2026-11-11T09:16:33.000Z","text":"I have been using llama.app, which is the desktop interface built on top of llama.cpp by the same people. It is an alternative to Ollama that might interest people who want to switch given Ollama's reputation and previous decisions. Llama.app has a really clean design, an elegant and intuitive interface. But it's good to get alternatives as well, if Apple also gets into the game.\nhttps://llama.app/","parent_id":50446520,"replies_count":2,"hn_url":"https://news.ycombinator.com/item?id=50451240"},{"id":50470124,"author":"dwaite","created_at":"2026-11-12T16:39:01.000Z","text":"By \"Apple\" they mean \"Claris\", a wholly owned subsidiary which was formerly Filemaker, Inc.\nSo this is interesting, but it will likely be an independent product rather than part of the core operating system.","parent_id":50446520,"replies_count":0,"hn_url":"https://news.ycombinator.com/item?id=50470124"},{"id":50453287,"author":"iantpe","created_at":"2026-11-11T13:39:16.000Z","text":"What I need most right now is a model like TypeSafe's Jev that can be built directly into the computer.\nJev allows for numerous small judgments: Is this email important? Which program should handle this request? Is this operation risky? There's no need for large LLMs to slowly generate long paragraphs of text; all you need is a fast, structured result that returns the message.\nHaving a local Jev-like model in the operating system for quick decision-making, and then calling the larger model when deeper thinking is needed, would be very useful.","parent_id":50446520,"replies_count":5,"hn_url":"https://news.ycombinator.com/item?id=50453287"},{"id":50451688,"author":"gehsty","created_at":"2026-11-11T10:14:21.000Z","text":"As models improve and hardware becomes more capable Apple are definitely well positioned.\nI am super interested to see what the Apple silicon team can deliver when their full attention is turned to local LLM acceleration.","parent_id":50446520,"replies_count":3,"hn_url":"https://news.ycombinator.com/item?id=50451688"},{"id":50451098,"author":"jwr","created_at":"2026-11-11T08:51:23.000Z","text":"I've been running local models under ollama, then lm-studio, and now mtplx (command-line, from supervisord). I'd welcome a vendor-supported solution, if it doesn't turn into bloatware (like lm-studio did).","parent_id":50446520,"replies_count":1,"hn_url":"https://news.ycombinator.com/item?id=50451098"},{"id":50447674,"author":"jedbrooke","created_at":"2026-11-10T23:16:17.000Z","text":"> Cluster for scale: “Connect two or more Thunderbolt 5-capable Mac computers and Harbor pools their unified memory into a single inference resource. Run models that exceed any single computer’s capacity, or serve more concurrent requests across your cluster.”\nThis seems like the main way it would differentiate itself over existing solutions like LMStudio etc.","parent_id":50446520,"replies_count":2,"hn_url":"https://news.ycombinator.com/item?id=50447674"},{"id":50448433,"author":"apparent","created_at":"2026-11-11T00:48:22.000Z","text":"This is what I've wanted for the last few years. I have played around with local models and had some success, but I don't want to deal with the hassle of setting up, updating, etc.\nIf all of this were easy, I would happily invest in new dedicated hardware to run it. This could be a full-blown Mac or it could be a souped up HomePod-type device, loaded with RAM. Better yet, make it a revitalized Matter-enabled Airport!","parent_id":50446520,"replies_count":1,"hn_url":"https://news.ycombinator.com/item?id=50448433"},{"id":50452602,"author":"KolibriFly","created_at":"2026-11-11T12:24:03.000Z","text":"Calculate the total cost of two new Mac Studios with max memory and cables. For that budget you could burn tokens on cloud APIs for a long time or build a dedicated server on open hardware with clear server buses","parent_id":50446520,"replies_count":5,"hn_url":"https://news.ycombinator.com/item?id=50452602"},{"id":50447737,"author":"gizmodo59","created_at":"2026-11-10T23:22:38.000Z","text":"I really want capable models local but it’s impossible for me to afford the hardware. If there is a magical way to run glm 5.3 without quantization at 80-100tps with a total spend of less than 10K I’d be happy but that’s impossible now.","parent_id":50446520,"replies_count":2,"hn_url":"https://news.ycombinator.com/item?id=50447737"},{"id":50451956,"author":"_pdp_","created_at":"2026-11-11T10:57:55.000Z","text":"That will be really cool. It is already possible though through Apple Intelligence. There is an open source app that already leverages that.\nhttps://github.com/pdparchitect/noodle","parent_id":50446520,"replies_count":0,"hn_url":"https://news.ycombinator.com/item?id=50451956"},{"id":50447157,"author":"mgiampapa","created_at":"2026-11-10T22:24:33.000Z","text":"I love that they are bringing Claris back for this. Moof!","parent_id":50446520,"replies_count":1,"hn_url":"https://news.ycombinator.com/item?id=50447157"},{"id":50451424,"author":"crossroadsguy","created_at":"2026-11-11T09:40:41.000Z","text":"I hope it will be optional. This is too much to ask for on Apple platform. I recently had to install a 3rd party script to get rid of Apple's laregely useless AI stuff from my 5 year old mac and save more than a few GBs which I badly needed.\nRunning local models is already extremely easy.\nI have a suspicion this app will be a way to add a too many gigabytes to the mac and of course like the Chess.app there will be no way to get rid of it. There are excellent cli apps/GUI apps like this https://github.com/ddalcu/mlx-serve - they are both cli and gui – and even its gui surprisingly nimble (I use this, but I am not related to the project). I am yet to try this https://github.com/ggml-org/Llama-macOS but I've seen only good reviews of it.\nI also do not think Apple is doing it for the users who won't explore around. I can't wait to get off this operating system. Next time a friend travels to the continent I will explore getting a Framework laptop via them.","parent_id":50446520,"replies_count":1,"hn_url":"https://news.ycombinator.com/item?id=50451424"},{"id":50450397,"author":"brainless","created_at":"2026-11-11T06:48:37.000Z","text":"I really do not understand the value of running large models locally. I am absolutely fond of local models but for specific tasks. The agents need to do a lot of work in this case: orchestration, big or small model split of planning and work, split into small tasks, merging, etc.\nLLMs with <30B parameter models seem like the sweet spot. I would rather focus on multiple tiny LLMs that are tuned to specific tasks (SQL queries, finance/accounting, backend/frontend dev, etc.).\nFor larger models, a $10-40 ish Neuralwatt/Opencode Go or any similar subscription goes so far for the average user that it is ridiculous. We are still in the \"youth\" era of agents. There are many patterns that work well with smaller LLMs. Or with a big-small LLM combination.\nMy context:\nI own the tiny 16GB M4 Mac Mini and I use LLMs (coding agents) to build software and a few of my products use LLMs directly. Full-time user of LLMs, do not read or write code manually. I run the tiny Qwen series (<10B params) sometimes but a couple $20 subscriptions (Claude, Neuralwatt) take care of my full-time agent usage.","parent_id":50446520,"replies_count":8,"hn_url":"https://news.ycombinator.com/item?id=50450397"},{"id":50447273,"author":"MBCook","created_at":"2026-11-10T22:35:17.000Z","text":"But you’re still in for a ton of pain if you want to run them headless right?\nThey desperately need to fix that.","parent_id":50446520,"replies_count":2,"hn_url":"https://news.ycombinator.com/item?id=50447273"},{"id":50454024,"author":"mark_l_watson","created_at":"2026-11-11T14:41:46.000Z","text":"I am on the waiting list. I have a new 64G mac mini and an older 32G model. The directions for setting up a distributed inference system aren’t too gnarly but I haven’t bothered. I will when I get access to Harbor.\nI think it is a great joy to run most things locally, and still occasionally use the best commercial models when local will not cut it.","parent_id":50446520,"replies_count":0,"hn_url":"https://news.ycombinator.com/item?id=50454024"},{"id":50454752,"author":"ontouchstart","created_at":"2026-11-11T15:44:37.000Z","text":"I am already doing 100% offline inferences and AI assisted coding with bespoke software developed with local models.\nIn my case it reached the recursive self improvement stage a few weeks ago. It feels like the early days of personal computing in the era of DOS and Apple II.\nThe hardware is already there. The only limit is product imagination.","parent_id":50446520,"replies_count":1,"hn_url":"https://news.ycombinator.com/item?id=50454752"},{"id":50461138,"author":"caycep","created_at":"2026-11-12T00:27:43.000Z","text":"Also I keep forgetting about Claris; I'm kind of sentimental about it and I hope maybe this gives them a shot in the arm.","parent_id":50446520,"replies_count":0,"hn_url":"https://news.ycombinator.com/item?id=50461138"}]}