15:[[47559143,{"by":"whatarethembits","descendants":15,"id":47559143,"kids":[47570951,47567968,47562218,47567988,47577562,47564409,47574479,47559625,47579548,47692010,47563142,47597799,47600225,47560852],"score":13,"text":"$16","time":1774742274,"title":"Ask HN: What's the latest concensus on OpenAI vs. Anthropic $20/month tier?","type":"story"}],[47597181,{"by":"aegis_camera","descendants":2,"id":47597181,"kids":[47597749,47597182],"score":1,"time":1775021615,"title":"Show HN: SwiftLM – Qwen Chat on iPhone, 100B+ Moe on M5 Pro 64GB (Native Swift)","type":"story","url":"https://github.com/SharpAI/SwiftLM"}],[47590389,{"by":"theseidel","descendants":9,"id":47590389,"kids":[47598921,47597736,47592800,47594430,47598908,47590403],"score":14,"time":1774976789,"title":"Show HN: Margo – Find the font your brain reads fastest","type":"story","url":"https://margo.fyi/"}],[47589420,{"by":"sankalpnarula","descendants":6,"id":47589420,"kids":[47597717,47599363,47599187],"score":5,"time":1774973023,"title":"Show HN: I built a self-hosted Fly.io engine using Go and Firecracker","type":"story","url":"https://github.com/herd-core/herd"}],[47588324,{"by":"izaidi","descendants":26,"id":47588324,"kids":[47589302,47590842,47625114,47600060,47607760,47599404,47591013,47604352,47600452,47594713,47598521,47595872,47617318,47589123,47597698,47604129,47593599,47588508,47588390,47609526,47602130,47591631,47595967,47596498,47593217],"score":66,"time":1774969217,"title":"Show HN: Sundial – a new way to look at a weather forecast","type":"story","url":"https://sundial.page/"}],[47566687,{"by":"tpurves","descendants":12,"id":47566687,"kids":[47581931,47570427,47568864,47579141,47568986,47573404,47585198,47567627,47692025,47609876,47576054,47596471,47567202],"score":11,"text":"To anyone upgrading their daily driver Mac this year, are you considering going to a Max + high memory config? eg. with the hope (now or in near future) of being able to do usefully run agents/LLMs locally on your main machine?

Or is the few extra thousand dollars difference between a base and max-spec MBP still just better spent on literally any other practical option (like different harware, remote hardware, cloud AI subscriptions or credits). Or wait to see if there will be an M5 Studio or what inferencing performance next year's M6 may bring?

I am tempted, but even with some new models getting skinnier and more efficient, I am not sure moore's law and the M5 generation is quite there yet to be worth the trouble?

What call's are ya'll making and why?","time":1774814578,"title":"Ask HN: M5 MacBook Pro buyers, worth spending the $$$ to maybe run LLMs local?","type":"story"}],[47582539,{"by":"elC0mpa","descendants":8,"id":47582539,"kids":[47593200,47583059,47587845,47585163,47595316,47588415,47589903,47605647],"score":6,"text":"Hey guys, I have a question related to these agentic clis

I work full time as a Software Engineer and recently started using tools like Claude Code and Gemini CLI. I just want to know your opinion about these because from y point of view they both have different disadvantages:

Gemini CLI usually takes long (~5ms) in plan mode because it uses Gemini Pro 3.1, but I have never been running out of tokens. However, Claude is pretty fast but I usually run out of tokens.

Just want to know if any of you have some workaround for any of these tools or a way to solve this.

Thank you!!!!","time":1774928951,"title":"Ask HN: Gemini CLI vs. Claude Code","type":"story"}],[47584908,{"by":"HaixuanTao","descendants":1,"id":47584908,"kids":[47585108],"score":1,"text":"Realtime Livestream Robot demo! ( for the next ~10h )\nhttps://huggingface.co/spaces/haixuantao/openarm

You can prompt the robot directly using \n@robot\n and the real robot is going to do the task!

You can also ask me to do stuff and i'll try my best!

More info at: https://x.com/HaixuanT/status/2038914443505983665?s=20","time":1774950537,"title":"Show HN: Livestream Robot Cooking","type":"story","url":"https://huggingface.co/spaces/haixuantao/openarm"}],[47571401,{"by":"secretdark","descendants":22,"id":47571401,"kids":[47573083,47571402,47572991,47575634,47571418,47573142,47589364,47576531,47573955,47573785,47573599,47575740,47579776,47585058,47574670,47575225,47572405,47576428,47577486],"score":40,"time":1774855160,"title":"Show HN: The Alphabetical Clock","type":"story","url":"https://boat.horse/clock/"}],[47582537,{"by":"markfront","descendants":1,"id":47582537,"kids":[47585041,47644547,47582538],"score":7,"time":1774928937,"title":"Show HN: Free AI API gateway that auto-fails over Gemini, Groq, Mistral, etc.","type":"story","url":"https://github.com/msmarkgu/RelayFreeLLM"}]]

user:	as1as
created:	March 31, 2026
karma:	3
about:	Solo developer building TalkWith.chat — daily AI debate battles with evolving personal AI debaters. Currently 0 users. Sharing the full build journey (mistakes included).