Podcasts about MacBook

You're missing out if you haven't explored Apple's accessibility menu. Mikah spotlights simple tweaks that solve real problems for everyone, not just those who need them. These are features you can toggle once, and love forever! Reducing car sickness with Vehicle Motion Cues and customizable dots Flash for Alerts: flash notifications with LED or screen for incoming calls and texts Custom haptic vibrations for messages and contacts Built-in white noise machine: Background Sounds on iPhone, iPad, and Mac Music Haptics: feeling rhythms and beats through the Taptic Engine Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord. Sponsor: threatlocker.com/twit

Dave Rubin of "The Rubin Report" gives a first look at the stories you need to know to start your day, including FBI agents seizing former Democratic congressman Eric Swalwell's cellphone, MacBook, and other electronics at San Francisco International Airport as federal authorities investigate multiple sexual-misconduct allegations that Swalwell denies; a Media Research Center study finding that Apple News and Google News overwhelmingly promote outlets with decades of Democratic presidential endorsements; and astrophysicist and UFO whistleblower Dr. Eric Davis claiming the U.S. government possesses spacecraft and remains from four alien species as he asks President Trump to waive an intelligence agreement so he can reveal classified information; and much more. #rubinreport #ericswalwell #mediabias #applenews #googlenews #ufo #aliens #daverubin

The news cycle is ending the week with absolute chaos. An alleged ISIS supporter is accused of plotting an attack on the New York State Capitol, while Scott Jennings' rumored future at the White House has Hasan Piker threatening legal action.Meanwhile, the drama surrounding Candace Owens and Turning Point USA continues to escalate. Andrew Wilson questions Charlie Kirk Show hosts about Candace, a TPUSA host reveals details about a surprising private conversation, and Candace says she doesn't care if Turning Point USA “burns down.”Plus, reports surrounding an FBI seizure of Eric Swalwell's phone and MacBook bring renewed attention to the California congressman, Michael Cohen makes surprising new claims about the prosecution of Donald Trump, and Jill Biden revisits Joe Biden's disastrous debate performance as a bizarre new theory emerges about what happened that night.The Chicks also cover James O'Keefe testing Alaska's voter ID rules, a Seattle Times columnist resigning after his girls' sports column was rejected, a Colorado teacher reprimanded for wearing a Protect Women's Sports shirt, the growing socialist influence inside Democratic politics, Lindsay Clancy's disturbing online fan movement, California's latest regulations, and an incredible happy ending for a Hamas hostage survivor.SUPPORT OUR SPONSORS TO SUPPORT OUR SHOW!Give your liver the support it deserves with Dose. Visit https://DoseDaily.co/CHICKS35 and use promo code CHICKS35 to get 35% off your first subscription.Get your summer glow-up with a skincare upgrade from Bon Charge. Visit https://BonCharge.com/Chicks and use code CHICKS for 15% off sitewideIf you're ready to stop feeling frustrated and start making a difference, download Concerned Women For America's FREE Biblical Civic Engagement Guide today at https://ConcernedWomen.org/ChicksSubscribe and stay tuned for new episodes every weekday!Follow us here for more daily clips, updates, and commentary:YoutubeFacebookInstagramTikTokXLocalsMore InfoWebsite

Apple accidentally leaked identifying numbers for much of its upcoming product line, AirPods with cameras rumors spark a new privacy debate, and Siri AI gets partnered articles on the AppleInsider Podcast.Please note that there's no video this week because of technical problems and pressing the wrong buttons. Video will be back next time.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:MasterClass: Get 15% off annual memberships at MasterClass.comLinks from the Show:macOS Tahoe 26.7 beta references several unreleased products like Home HubWatch how AirPods with cameras will work, according to macOS TahoeThere is a good reason to praise AirPods with cameras, and not fear themWhat are the rumors about the AirPods Pro 4?A new Siri Remote may be in the works, but don't get excitedApple's product identifier leak is a hard mystery to solveSiri AI could get a boost from news content with new publisher dealApple's first foldable iPhone won't arrive everywhere in SeptemberApp Store review is broken in a time where it is needed the mostApple's latest commission rates for external App Store purchases haven't satisfied EpicEpic doesn't believe in compromise unless it is with GoogleEpic Games vs Apple -- The continuing six-year App Store sagaApple's EU fee changes, which could eventually hit the USJony Ive-designed Ferrari Luce sells for $40M at auctionIt's begun - US and Canadian firms can now buy ads on Apple MapsApple Maps ads will be tasteful and minimal, might even help your local businessesSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Intro (11:59) - AirPods Pro with cameras (32:56) - Siri AI and publishers (45:41) - App Store problems (56:12) - Epic ★ Support this podcast on Patreon ★

Confused by the Apple Watch's shifting gestures and button functions? Mikah Sargent breaks down the new logic, revealing how a simple mantra can help you master your device's next overhaul. watchOS 27 overhauls Apple Watch button functions and navigation Breaking old muscle memory: changes to Digital Crown, side button, Siri Digital Crown press reveals new dynamic app switcher and app list Scrolling Digital Crown now brings up the Smart Stack of widgets Side button revamps: Control Center, Apple Pay, Emergency SOS/Power Dynamic app grid's recent and context-based app selection Updated Apple Watch gestures: double tap, single tap, wrist flick for navigation Quick mantra: press for apps, turn for widgets, side for control, hold for Siri Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

FROM ROLLING STONE: Jonathan Richman moved from Boston to New York as a teenager in hopes of sleeping on Lou Reed's couch. That influence shows on the two-chord anthem “Roadrunner.” Recorded in 1972 but not released until 1976, Lovers turned the tough sounds of the Velvets into an ode to suburban romanticism. “Rock & roll was about stuff that was natural,” Richman said. “I wasn't about drugs and space.” Songs like “Pablo Picasso,” “Girl Friend,” and “Dignified and Old” touched generations of punk and indie-rock innocents. Note From Rob: Some of the audio for the sfx and music sucks because my computer sucks. How can a MacBook bought this year not handle RECORDING A PODCAST? I am too lazy to fix it all, and the music is not that great anyway. 

Elapsed Time Adder is Live in the App Store! MacBook Neo – Can It Replace an iPad mini? Why I Left Fantastical and Todoist for Structured ETENWOLF Zephyr S3 Air Compressor Support the Show Apple Photos – Making Your Memories Last with an Easy to Access Archive Audiobook Audio (V3) Part 7: The Noise You Notice Visualising Trails with Google Earth Pro Transcript of NC_2026_08_16 Join the Conversation: allison@podfeet.com podfeet.com/slack Support the Show: Patreon Donation Apple Pay or Credit Card one-time donation PayPal one-time donation Podfeet Podcasts Mugs at Zazzle NosillaCast 20th Anniversary Shirts Referral Links: Setapp - 1 month free for you and me 15% off Carbon Copy Cloner Wispr Flow - 1 month free for you PETLIBRO - 30% off for you and me Parallels Toolbox - 3 months free for you and me Learn through MacSparky Field Guides - 15% off for you and me Backblaze - One free month for me and you Eufy - $40 for me if you spend $200. Sadly nothing in it for you. PIA VPN - One month added to Paid Accounts for both of us CleanShot X - Earns me $25%, sorry nothing in it for you but my gratitude

The iPhone 18 Pro will be a powerhouse for photography while the iPhone Ultra will focus on the novelty of folding, your hosts discuss this and the AI competition in the latest AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:CleanMyMac by MacPaw: Get Tidy Today! Try 7 days free and use  code APPLEINSIDER20 for 20% off at clnmy.com/APPLEINSIDERLinks from the Show:iPhone 18 Pro rumor roundup: 2nm A20 chip, C2 modem, under-display Face IDPhoto leak backs up massive iPhone 18 Pro Max battery rumoriPhone 18 Pro expected to be 40% more expensive to make than iPhone 17 ProEveryone at Apple is calling the foldable 'iPhone Ultra'Jefferies cuts AAPL target to $263.66, downgrades to UnderperformFirst hints emerge about how much extra Apple Intelligence features will costOpenAI's new hardware leaks: Over $300, same size as an Alexa DotNew ChatGPT version has a 'Think' button, will find 'more reliable facts'RAM shortages now affecting MacBook Air availabilityGenerative AI needs ethical leaders and to be controlled like nuclear powerSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - iPhone 18 Pro (15:12) - iPhone Bendy (17:20) - Jeffries Tube (24:51) - Apple Intelligence cost and cameras (36:04) - OpenAI's HomePod (45:24) - RAM shortages ★ Support this podcast on Patreon ★

Stop fumbling with serial numbers, plant names, or care tags. Discover quick, practical tricks to turn your iPhone's camera into a powerhouse for instant web answers and hassle-free text capture. Enabling Live Text and checking device and language compatibility Using Live Text in Photos, Camera, and Notes for text selection Tips for extracting phone numbers, emails, and addresses with Live Text Scanning serial numbers and copying image text across Apple devices Identifying plants, animals, landmarks, and more with Visual Lookup Visual Lookup for laundry symbols and car dashboard lights Troubleshooting fussy features Best practices for clear OCR and uncluttered frames Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Stop fumbling with serial numbers, plant names, or care tags. Discover quick, practical tricks to turn your iPhone's camera into a powerhouse for instant web answers and hassle-free text capture. Enabling Live Text and checking device and language compatibility Using Live Text in Photos, Camera, and Notes for text selection Tips for extracting phone numbers, emails, and addresses with Live Text Scanning serial numbers and copying image text across Apple devices Identifying plants, animals, landmarks, and more with Visual Lookup Visual Lookup for laundry symbols and car dashboard lights Troubleshooting fussy features Best practices for clear OCR and uncluttered frames Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Stop fumbling with serial numbers, plant names, or care tags. Discover quick, practical tricks to turn your iPhone's camera into a powerhouse for instant web answers and hassle-free text capture. Enabling Live Text and checking device and language compatibility Using Live Text in Photos, Camera, and Notes for text selection Tips for extracting phone numbers, emails, and addresses with Live Text Scanning serial numbers and copying image text across Apple devices Identifying plants, animals, landmarks, and more with Visual Lookup Visual Lookup for laundry symbols and car dashboard lights Troubleshooting fussy features Best practices for clear OCR and uncluttered frames Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Stop fumbling with serial numbers, plant names, or care tags. Discover quick, practical tricks to turn your iPhone's camera into a powerhouse for instant web answers and hassle-free text capture. Enabling Live Text and checking device and language compatibility Using Live Text in Photos, Camera, and Notes for text selection Tips for extracting phone numbers, emails, and addresses with Live Text Scanning serial numbers and copying image text across Apple devices Identifying plants, animals, landmarks, and more with Visual Lookup Visual Lookup for laundry symbols and car dashboard lights Troubleshooting fussy features Best practices for clear OCR and uncluttered frames Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Stop fumbling with serial numbers, plant names, or care tags. Discover quick, practical tricks to turn your iPhone's camera into a powerhouse for instant web answers and hassle-free text capture. Enabling Live Text and checking device and language compatibility Using Live Text in Photos, Camera, and Notes for text selection Tips for extracting phone numbers, emails, and addresses with Live Text Scanning serial numbers and copying image text across Apple devices Identifying plants, animals, landmarks, and more with Visual Lookup Visual Lookup for laundry symbols and car dashboard lights Troubleshooting fussy features Best practices for clear OCR and uncluttered frames Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

A.M. Edition for Aug. 10. Iran dials up its demands in talks to reopen the Strait of Hormuz. Plus, Meta embraces open-weight AI models in a bid to blunt the appeal of cheaper Chinese competition. And WSJ finance editor Alex Frangos breaks down how a payment network backed by the Russian government reveals the limits of Western efforts to economically isolate Moscow. Luke Vargas hosts. Sign up for the WSJ's free What's News newsletter. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.

Ki rendeli a pizzát? Ki kívánja meg a dobostortát? Ki hozza a döntéseket bennünk? Ebből az adásból kiderül, hogy erre a kérdésre már nem lehet rávágni azt, hogy a főnök a fejünkben ül: Kékesi Balázs filozófus, akit sokan Bauxitként ismernek a Bëlga zenekarból, legalábbis arra világít rá, hogy más-más módon, de ma már több tudományterület is megfogalmazza, hogy olykor egy egész bizottság vagy ha úgy tetszik, egy egész erdő dönt helyettünk. Vagyis ebben az epizódban Kékesi Balázzsal beszélgetek kognitív nyelvészetről, durva vagy csak elemzésre sarkalló dalszövegekről, a szokások hatalmáról, a kilencvenes évek buddhista főiskolájáról, kényszerképzetekről és megszabadulásról, az életről meg mindenről.

Un nuevo APPLEaks cargadísimo de rumores y polémicas.Apple estaría preparando gafas con inteligencia artificial, nuevos AirPods Ultra con cámaras, el esperado iPhone plegable y cambios importantes para los iPhone 18 Pro y Pro Max.Además, analizamos el posible problema de disponibilidad del iPhone 18 Pro, las mejoras de cámara y batería, los rumores de subida de precio y el regreso de una idea que parecía descartada: un nuevo MacBook Ultra con pantalla táctil.Y también hablamos del misterioso dispositivo de OpenAI y Jony Ive que podría convertirse en uno de los grandes rivales del ecosistema Apple.¿Cuál de todas estas novedades te parece más importante?00:00 Apple prepara gafas con inteligencia artificial01:35 AirPods Ultra con cámaras: ¿para qué servirían?03:23 El misterioso dispositivo de OpenAI y Jony Ive05:19 iPhone plegable: el gran problema de la pantalla07:44 ¿Habrá escasez del iPhone 18 Pro?09:41 Apple domina el mercado premium10:42 Apple habría abandonado el modo Cine11:42 Las grandes novedades del iPhone 18 Pro12:09 La mayor batería de la historia12:34 El iPhone 18 Pro sería más caro13:34 Lo próximo de Apple para 202714:05 ¿MacBook Ultra con pantalla táctil?Apple, AppleLeaks, iPhone 18, iPhone 18 Pro, iPhone 18 Pro Max, iPhone Fold, iPhone plegable, Apple Fold, AirPods Ultra, Apple Glasses, gafas Apple, Apple Intelligence, inteligencia artificial Apple, MacBook Ultra, MacBook Pro, MacBook Air M6, MacBook M6, OpenAI, Jony Ive, iPhone 18 rumores, iPhone 18 precio, iPhone 18 cámara, iPhone 18 batería, Apple septiembre 2026, evento Apple, nuevos productos Apple, noticias Apple, rumores Apple, idearVlog#Apple #iPhone18Pro #APPLEaks

Pendant les vacances, Sortie de veille change de formule ! À la place de notre traditionnel résumé hebdomadaire de l'actualité, nous réunissons plusieurs membres de l'équipe pour débattre de sujets qui fâchent. L'occasion de vous faire vivre les discussions qui animent au quotidien la rédaction de MacGeneration.Pour ce troisième hors-série « opinion impopulaire », Cédric défend l'idée que le retour du connecteur MagSafe sur Mac n'était pas nécessaire. À quoi bon avoir un connecteur propriétaire qui ne fait qu'une chose quand l'USB-C permet déjà de recharger les MacBook, et ce à moindre coût ? Christophe, Pierre et Stéphane n'adhèrent pas vraiment à cet avis… Hébergé par Acast. Visitez acast.com/privacy pour plus d'informations.

Someone has to pay for all this AI, nothing is foolproof, and Apple's lawsuit with OpenAI is only just beginning, all on the AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechLinks from the Show:Apple AI compute costs will be covered by iCloud+ subscriptions, for nowMeta's Muse Code is yet another AI coding agent on macOSRAM production worldwide is sold out through 2027The world's RAM supply crisis is going to get worse before it gets betterWhy you should never put a hot iPhone in the fridge or freezerAnother iPhone 17 Pro plummets from plane, and found intactTilta Khronos iPhone 17 Pro Max photography kit reviewPSA: don't fall for the scam Uber email making the roundsGoogle Health now shares Fitbit workouts with Apple HealthWebKit leaks in iOS & macOS expose user data in spite of proxy useNew OpenAI post doesn't address Apple's IP theft claims at allConfidential Apple files followed former employees to OpenAI through iCloudApple demands OpenAI injunction, discovery, testimony now to prevent more harmApple's corporate espionage suit against OpenAI: How we got hereTelegram's takedown caused by planted porn & weaponized App Store policiesTelegram briefly disappeared from the Apple App Store due to CSAM2027 iPhone 18 rumors: Smaller Dynamic Island, 12GB or 9GB RAM, 2nm A20 chipiPhone Fold is on schedule for September after manufacturing problems are solvedStaff lottery reveals Apple's first-half September event plansApple's iPhone 18 Pro event won't be live, no matter what you've heard todaySupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Apple Intelligence tokens (19:35) - RAM crisis (30:41) - Do not freeze your hot iPhone (39:26) - Don't drop iPhones from planes (44:53) - Email scams, Google Health, and OpenAI ★ Support this podcast on Patreon ★

If you have kids using iPhones, Screen Distance might already be enforcing healthy habits without you knowing it. Find out how iOS screen features, like Bold Text and True Tone, are shaping digital wellness for your eyes and your family. Setting up and using Screen Distance on Face ID devices Night Shift: color temperature adjustments and sleep science skepticism True Tone vs. Night Shift: real-time white balance adaptation Additional display tweaks: Auto-Brightness, Reduce White Point, Larger Text Tracking outdoor time with Apple Watch's Time in Daylight Outdoor time as a proven method to reduce myopia risk Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

If you have kids using iPhones, Screen Distance might already be enforcing healthy habits without you knowing it. Find out how iOS screen features, like Bold Text and True Tone, are shaping digital wellness for your eyes and your family. Setting up and using Screen Distance on Face ID devices Night Shift: color temperature adjustments and sleep science skepticism True Tone vs. Night Shift: real-time white balance adaptation Additional display tweaks: Auto-Brightness, Reduce White Point, Larger Text Tracking outdoor time with Apple Watch's Time in Daylight Outdoor time as a proven method to reduce myopia risk Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

If you have kids using iPhones, Screen Distance might already be enforcing healthy habits without you knowing it. Find out how iOS screen features, like Bold Text and True Tone, are shaping digital wellness for your eyes and your family. Setting up and using Screen Distance on Face ID devices Night Shift: color temperature adjustments and sleep science skepticism True Tone vs. Night Shift: real-time white balance adaptation Additional display tweaks: Auto-Brightness, Reduce White Point, Larger Text Tracking outdoor time with Apple Watch's Time in Daylight Outdoor time as a proven method to reduce myopia risk Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Do smart glasses need video recording to be relevant? The MacVoices Live! panel debate the importance of privacy, accessibility, navigation, object recognition, captions, and visual assistance as measures of usefulness. They examine the stigma created by existing smart glasses, the difference between processing and storing images and the public expectations around recording, detection apps, and hidden cameras. The panel includes: Chuck Joiner, Web Bixby, Eric Bolden, Marty Jencius, David Ginsburg, Brian Flanigan-Arthurs, Jim Rea, Jeff Gamet, and Mark Fuccio.  This edition of MacVoices is brought to you by our Patreon supporters. Get access to the MacVoices Slack and MacVoices After Dark by joining in at Patreon.com/macvoices. Show Notes: Chapters: 00:00 Do Apple glasses need video recording? 02:08 Smart glasses as visual-assistance devices 03:37 The surveillance stigma created by existing products 04:26 Cameras, environmental processing, and privacy 05:53 Processing images without permanently recording them 07:31 Wearing smart glasses in theaters, meetings, and private spaces 09:03 Accessibility, captions, navigation, and low-vision assistance 10:32 Evaluating a product Apple has not yet announced 13:01 Apps that detect nearby smart glasses 13:59 How often people use cameras on their devices 15:14 Scanning documents and scrapbooks with an iPad 17:01 Comparing iPad, iPhone, and MacBook cameras 17:13 Smart-glasses detection apps and Bluetooth scanning 19:05 What should happen when recording glasses are detected? 19:40 Prescription lenses and expectations of privacy 21:03 Tiny cameras and unavoidable modern surveillance 21:59 Early sketches and prototypes of the Finder icon Links: Apple Glasses just won't be useful without video recording - 9to5Mac https://9to5mac.com/2026/07/27/apple-glasses-just-wont-be-useful-without-video-recording/   Antizuck Smart Glasses Scanner App - App Store https://apps.apple.com/us/app/antizuck-smart-glasses-scanner/id6785557805 Guests: Get detailed bios and contact information about for the panel on the MacVoices Live! Panel page on our web site: https://macvoices.com/macvoiceslive/macvoices-live-panel/ Support:      Become a MacVoices Patron on Patreon      http://patreon.com/macvoices      Enjoy this episode? Make a one-time donation with PayPal Connect:      Web:      http://macvoices.com      Twitter:      http://www.twitter.com/chuckjoiner      http://www.twitter.com/macvoices      Mastodon:      https://mastodon.cloud/@chuckjoiner      Facebook:      http://www.facebook.com/chuck.joiner      MacVoices Page on Facebook:      http://www.facebook.com/macvoices/      MacVoices Group on Facebook:      http://www.facebook.com/groups/macvoice      LinkedIn:      https://www.linkedin.com/in/chuckjoiner/      Instagram:      https://www.instagram.com/chuckjoiner/ Subscribe:      Audio in iTunes      Video in iTunes      Subscribe manually via iTunes or any podcatcher:      Audio: http://www.macvoices.com/rss/macvoicesrss      Video: http://www.macvoices.com/rss/macvoicesvideorss

If you have kids using iPhones, Screen Distance might already be enforcing healthy habits without you knowing it. Find out how iOS screen features, like Bold Text and True Tone, are shaping digital wellness for your eyes and your family. Setting up and using Screen Distance on Face ID devices Night Shift: color temperature adjustments and sleep science skepticism True Tone vs. Night Shift: real-time white balance adaptation Additional display tweaks: Auto-Brightness, Reduce White Point, Larger Text Tracking outdoor time with Apple Watch's Time in Daylight Outdoor time as a proven method to reduce myopia risk Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

If you have kids using iPhones, Screen Distance might already be enforcing healthy habits without you knowing it. Find out how iOS screen features, like Bold Text and True Tone, are shaping digital wellness for your eyes and your family. Setting up and using Screen Distance on Face ID devices Night Shift: color temperature adjustments and sleep science skepticism True Tone vs. Night Shift: real-time white balance adaptation Additional display tweaks: Auto-Brightness, Reduce White Point, Larger Text Tracking outdoor time with Apple Watch's Time in Daylight Outdoor time as a proven method to reduce myopia risk Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Watch The FULL EPISODE on Patreon!https://www.patreon.com/c/THEUNCUTPODCASTImportant: To avoid paying extra fees, please don't subscribe through the iOS App Store.Instead, subscribe via a PC, MacBook, or laptop using your web browser. This way, you'll avoid the additional charges applied through the App Store.Send us your dilemma here: https://uncutpodcast.komi.io.Follow us on our personal Instagram accounts:Beatrice - https://www.instagram.com/beatriceakn/Tammy - https://www.instagram.com/tammymontero/Sharon - https://www.instagram.com/sharonodu/ Hosted on Acast. See acast.com/privacy for more information.

Watch the full episode on YouTube:We first covered Baseten last year when DeepSeek mania was at peak hype. Now they have raised a monster $13B round and become one of the new cohort of AI Infra decacorns that are (with Nvidia, Intel, and the semis complex) chief beneficiaries of the Inference Inflection. We return to Baseten at the peak of the 2026 edition of Open Weights debate. Ali has published a viral breakdown of Kimi K3:And since you last saw him, Philip has spoken at AI Engineer and written the definitive book on Inference Engineering spotted all over SF:Three years ago, inference engineering barely existed as a category.Today, it is one of the most critical disciplines in AI. Inference engineering inherently tackles a different question than standard model training: “How do you turn those weights from training into a product that is fast, reliable, and affordable at scale?” Focusing on these creates an entirely new optimization problem.In one recent GLM-5.2 experiment, quantizing more of the model actually preserved its benchmark quality while increasing throughput by 20%, because the errors introduced in different layers could cancel each other out.Inference is no longer just the final step after training. It is becoming its own engineering discipline, with its own research problems, infrastructure, and increasingly specialized roles.In this episode, Baseten's Philip Kiely and Ali Taha join swyx and Vibhu to explain what actually happens after a new open model is released and what it takes to turn “we generated a token” into a fast, reliable, production-ready API.We go deep on cache-aware routing, disaggregated prefill and decode, quantization, speculative decoding, KV-cache movement, model parallelism, GPU kernels, and the race to make frontier models up to 10× faster. Philip and Ali explain why inference optimizations can still produce gains of 20%, 100%, or even 200%; how quantization errors can cancel one another out; why identical weights can behave differently across clusters; and how Baseten grafted a Kimi vision encoder onto GLM-5.2 without changing the underlying language model.The conversation then expands beyond LLMs into NVIDIA Dynamo, mega kernels, Rubin, AI-specific chips, local inference, video generation, diffusion versus autoregressive models, and the enormous compute barrier to generating coherent long-form video. Finally, we explore the convergence of training and inference, continual learning through persistent KV cache, and the emerging loop where models help optimize the infrastructure that runs them.We discuss:* What happens when a 200,000-token request enters an inference system* Cache-aware routing and reusing previously computed KV cache* Why prefill and decode are increasingly handled by different GPUs* When dedicated deployments become cheaper and more reliable than shared APIs* How speculative decoding uses a smaller model to accelerate a larger one* Tool calling, structured outputs, and what LLMs actually do* What it takes to support a new open model on day zero* Grafting Kimi's vision encoder onto GLM-5.2* Retrofitting inefficient model layers with components from other architectures* Why models sometimes collapse into repeating the same token* How hardware, kernels, and race conditions create nondeterministic failures* Preserving model fidelity while making inference faster* How quantization errors can cancel each other out* Why inference optimizations still deliver gains of 20%, 100%, and 200%* How optimized serving can make a model up to 10× faster* NVIDIA Dynamo, KV-aware routing, and distributed model serving* Speculative decoding the speculative decoder* Why local AI is about making models less dumb while data-center AI is about making them less slow* Tensor, expert, and pipeline parallelism across GPUs* Hardware-aware model design, auto-tuning, and the case against mega kernels* Rubin and why inference is becoming a systems problem* Whether modern GPUs are evolving into programmable AI ASICs* Why enormous models like Kimi K3 require GB300-class hardware* Why open-source video generation still trails Veo, Kling, and other closed models* The quadratic attention bottleneck behind long-form AI video* Autoregressive video, real-time generation, and compounding quality drift* Why future video systems may combine autoregressive and diffusion architectures* Training for inference and inference for training* Continuous post-training, deployment, evaluation, and improvement loops* How GLM-5.2 helped optimize the kernels serving GLM-5.2 itself* Why faster networking could unlock dramatically faster decoding* Continual learning, KV-cache compaction, and persistent model memoryShow Notes* How to build a day-0 API for Kimi K3* 22580: From GPT2 to Kimi3, ExplainedPhilip Kiely* LinkedIn: https://www.linkedin.com/in/philipkiely* X: https://x.com/philipkiely* Inference Engineering: https://www.baseten.co/inference-engineering/Ali Taha* LinkedIn: https://www.linkedin.com/in/aliestaha/* X: https://x.com/waterloointernTimestamps00:00:00 Introduction and the 200K-Token Prompt00:03:18 Dedicated Deployments, Speculative Decoding, and Tool Calling00:11:26 Launching Production-Ready Open Models00:19:06 Model Retrofits, Failure Modes, and Nondeterminism00:28:22 Quantization and Canceling Errors00:32:15 The Race to 10× Faster Inference00:40:48 Dynamo, Speculation, and Local vs. Data-Center AI00:50:18 Model Parallelism, Auto-Tuning, and Mega Kernels01:00:55 Rubin, GPUs vs. ASICs, and Custom AI Chips01:10:03 Giant Models and the Limits of GPU Memory01:12:42 AI Video, Quadratic Attention, and Autoregressive Generation01:21:47 Audio, Images, and Diffusion Models01:27:32 Training, Self-Optimizing Models, and Continual Learning01:40:06 Closing ThoughtsTranscriptIntroduction: Baseten, Waterloo Intern, and Inference EngineeringSwyx [00:00:00]: Okay, we're here in the studio with Philip, old friend from Inference Engineering, the book, as well as Baseten and everything that you've done, you and I have done before, as well as Ali. Welcome.Ali [00:00:15]: Pleasure to meet you.Swyx [00:00:15]: Waterloo intern.Ali [00:00:16]: Waterloo intern, always.Swyx [00:00:17]: When did you get “Waterloo intern” as a handle?Ali [00:00:19]: As a handle? Oh.Ali [00:00:20]: I think the rebranding happened mid-March. When I saw it was open, I was like, “I have to take it. Up for grabs.”Philip [00:00:26]: The problem is that Ali is really good at his job and is not gonna be an intern much longer.Philip [00:00:30]: So we have to figure out who's gonna get the handle.Ali [00:00:33]: Well, I'll pass the torch over to the next intern.Swyx [00:00:34]: Oh, okay. It can be, like, you just pass it to another Waterloo grad.Ali [00:00:37]: To another Waterloo intern. No, bruh.Philip [00:00:39]: Yeah.Ali [00:00:39]: Intern.Swyx [00:00:40]: Intern, yeah.Ali [00:00:40]: And no.Philip [00:00:41]: You gotta get an intern from Waterloo.Ali [00:00:42]: Yeah, I've gotta get an intern from Waterloo.Swyx [00:00:44]: Right.Ali [00:00:44]: But they have to follow the path.Swyx [00:00:45]: Oh, it could, but it could come from Baseten, so it's like whoever Baseten gets from Waterloo.Ali [00:00:48]: Right.Swyx [00:00:49]: Has the title of Waterloo.Ali [00:00:50]: It stays in the ecosystem.Philip [00:00:51]: Exactly.Ali [00:00:52]: Halfway through the internship, you either get it or you're out.Philip [00:00:55]: You should also do, like, a big graduation ceremony where you change the handle.Ali [00:00:59]: Just say it.Philip [00:00:59]: For everybody.Swyx [00:01:00]: You guys are good at ceremonies, clearly. We had a nice launch of the book, very successful. But before we get into all that, I wanna start off with a fun question for you. Okay, you're an expert inference engineer. What happens when I send a long query, say two hundred thousand tokens into Baseten's inference? What's the process of query through GPU model routing, balancing, all that? What is all the stuff that we don't think about?Long Context Requests, KV Cache, and Cache-Aware RoutingPhilip [00:01:26]: With a long query specifically, the first thing that I'm gonna ask is, “Have you sent me this query before, or at least part of it?” and I really hope you have, because it's gonna be a lot easier for me and a lot cheaper for you. So the first thing that we're gonna look at is some cache-aware routing, where we're going to see, we probably have a number of instances, a number of replicas up serving whatever model you're hitting. We want to send this one to something with, number one, available prefill workers, and number two, ideally some cached input already there so that we can skip prefill on at least part of these two hundred thousand tokens. If you're doing two hundred thousand tokens, it's probably coding or a multi-turn agent or something where you would expect to have that cached. If you don't, we're gonna have to send it to a prefill worker. We've at least on certain models disaggregated prefill and decode, so you're going to have one set of GPUs that's solely going to process the input, create the KV cache, and get you your first token, and then that's going to be passed over to a separate set of GPUs, which is going to run decode. We're going to iteratively make those tokens. We're probably going to have some speculator model in front of that. I'm going to assume that you're doing coding, and because of that, our speculator model, which assumes you're doing coding, is gonna have a high draft token acceptance rate. If I'm wrong and you're asking me to summarize every Harry Potter book, it's gonna be slower. And then we stream that output to you and account for it, charge you, a couple of pennies and say, “Hey, would you like to send another one?”Swyx [00:03:04]: Except Baseten doesn't charge by pennies.Philip [00:03:07]: Well, yeah, we charge. I'm assuming that we're talking about the public model APIs. If you are setting up a dedicated deployment, then yeah, it's not pennies.Public APIs vs. Dedicated DeploymentsSwyx [00:03:18]: Yeah, one of the key differentiators when I was talking with Baseten initially was that people who want very high volume just need to rent by the box, ‘cause then it's up to you to figure out how to saturate the box.Ali [00:03:31]: And more often than not, it's, like, way cheaper if you're pushing, like, millions of tokens per hour, if you just pay per hour instead of pay per token.Philip [00:03:37]: Yeah, they do. I think that we've increasingly seen a lot of demand for the pay per token APIs, just because everyone wants to try open models, and then once they find a use case that's really sticky, then they move over to dedicated.Swyx [00:03:51]: Is there a best practice on when it's time to swap over?Philip [00:03:54]: Couple reasons. Yeah, reliability, that's a big one, right?Ali [00:03:57]: Like, if they have a very specific use case, they want you to train something specifically for them, like they want their own spec dec, for instance, for their own traffic.Swyx [00:04:04]: Spec dec is speculative decoding.Speculative Decoding and Custom SpeculatorsAli [00:04:05]: Speculative decoding, yeah.Swyx [00:04:07]: You have to explain.Ali [00:04:07]: Sorry. Like, speculative decoding is like, if you have a huge model, right? And so the model is going to be generating one token at a time every single turn, every single forward pass. So we attach, like, this little, like, parasite, like this layer that goes on top of the model, and this model just has to predict. It does three very fast autoregressive forward passes, and it will predict, like, three certain tokens, and then you do one forward stage over the entire original model in order to see if those predictions were correct or not, and then you accept them or you reject them. Now, this draft model is traffic specific, so if you, like, Philip said, if you're summarizing Harry Potter books, I can train exclusively that draft model on Harry Potter books, and I can guarantee you that I'm gonna accept the three tokens every single time. And so with that case, I increase your decode speed. I wouldn't be able to provide this to you if you're a shared endpointSwyx [00:04:53]: YeahAli [00:04:53]: ‘cause I have no idea if you're doing Harry Potter, if you're doing coding, if you're doing English. We don't know. Also, there was a thing in the book that mentioned that if they really cared about a specific threshold, chapter four, I think. Do you remember that?Philip [00:05:06]: Yeah. The things that you can do is you can set a specific, like, batch sizing, a specific, like, parallelism strategy if you're trying to optimize for, like, throughput versus latency. You can. Maybe a NVFP4 quant doesn't pass your benchmarks and you wanna run a model at higher precision, you could do that. There's just a bunch of reasons why you might wanna have your own endpoint and the biggest one, of course, just being, like, you don't have to deal with someone else doing a hundred million tokens of benchmarking traffic at the endpoint when you happen to be trying to serve your users.Swyx [00:05:40]: Yeah. I think one thing that is. That is a classic journey. Like, it's people is asking the, what happens when you type Google into the browser. Tool calling, is that just, you're generating JSON or is there more complication beyond that?Tool Calling, JSON, and Structured OutputsAli [00:05:58]: Certain customers that we have, they have their own post-trained models, and so they demand a tool calling that's not just, like parse a file or go find the weather. It's something that's very specific and you have to do post-training on this. And if the post-training on the model is not good or if the quantization after the post-training to get the inference to be fast, the model will struggle reading the JSON file and reading the tool calling. But it doesn't require its own like sandbox. It's not like it's going to use that tool calling to like escape a sandbox or like it doesn't have to be contained. It can just be a normal dedicated deployment. The challenge with tool calling more and more seems to be that the companies want certain tool calling which is a very sensitive thing to train. And because you're dealing with all of the JSON outputs, if it doesn't like close the end of the request in a very certain manner, you end up with a model that did the tool calling and like the thinking and so as a result of that, it didn't see the result and just hallucinated the result as it decoded. That seems to be the most challenging thing with tool calling, not really the sandboxes model.Philip [00:06:56]: Yeah, that's a challenge on the training side and then on the inference side, there's work that you can do to scope the possible output. So we published this at this point close to two years ago, the solution to this problem which is you make a state machine and you use that to constrain the output to a specific format. So this is the structured output problem. If you remember backSwyx [00:07:27]: Yeah, the specific grammar is,Philip [00:07:29]: Yeah, exactlySwyx [00:07:30]: GML had this thing.Philip [00:07:31]: Yeah. So it's like the old-school “make sure this is only JSON”, return only JSON orSwyx [00:07:38]: YeahPhilip [00:07:38]: Grandma's gonna die type of prompts.Swyx [00:07:39]: Is it BNF grammar? At some point OpenAI had released a thing that was like, yeah, if you want to constrain your output, write BNF grammar, back as NOR.Philip [00:07:47]: In our inference system, it's just a specified output format. And you get the guarantee that your output's gonna be structured along that format. And so applying that to tool calls can like help cut down on. You can still call the wrong tool or call no tool. It doesn't solve the certainty problem but it at least solves the output structuring problemSwyx [00:08:10]: YeahPhilip [00:08:10]: Within tool calls.Swyx [00:08:12]: And MCP is just another form of tool, right.Philip [00:08:14]: Yeah, exactly.Swyx [00:08:15]: As far as there's no special thing there.Philip [00:08:16]: The thing I'm always like explaining to people is the LLM is not capable of doing anything. It's only capable of making suggestions of what to do and then if those suggestions are formatted in a certain way and applied to a system that knows what to do with them, then an action occurs.Swyx [00:08:32]: Yeah. Part of the fun stuff is, this is solved outside of tool calling too. Like in an agent loop if the output is not correct or you're right, like reasoning, tool calling was done in the reasoning trace, just be like, “Oh, I don't know what to do. Let me just try again.” And it might get there after a few tries. And on your point of training, sometimes this is harder in smaller models, so you don't have the same exact quality outputAli [00:08:56]: Right.Swyx [00:08:57]: When you just swap from a big model, right?Ali [00:08:59]: Yeah. I will say that, before, I think we need to go back to inference engineering proper.Ali [00:09:04]: But, I had expected that something would replace JSON because it's hard to stream JSON ‘cause JSON must be complete and you must have open and close brackets and everything. So it's hard to parse something or validate something while it's being streamed. So people invented all sorts of things that are like, I forget the name of some of these alternatives, but it's something like TOML, something like YAML. But JSON seems to be dominant still.Philip [00:09:30]: The JSON outputs aren't that long, right? Like you could have a long-- ‘cause tool calls also contain the arguments in them and perhaps for a certain tool you might pass like a very long argument. But my impression of the median tool call is that it's a relatively small number of tokens, right? So I would expect that speculators are generally fairly good at something as formatted as JSON. And so you would have like a pretty fast decode step there and that the streaming wouldn't be as valuable, but maybe I'm wrong about that.Ali [00:10:02]: I think you're also bounded by the software or that the model is gonna integrate with if the software is built with JSON for the tool calls or if the company that you'- if your customer says that this is how our software works and our tools are interfaced with JSON, you can ask them to like, change their software and say like, “Yeah, this is gonna be better for the model.” but like with the right training shouldn't be that much of a difference. Also more profitable if it outputs more tokens probably.Swyx [00:10:25]: Depends on your business model.Swyx [00:10:27]: It really depends. But I will say that, as a writer with like experience a lot with generated output, I do try to move from text to JSON text which is very long JSON, right? Like there's paragraphs in every field because I'm trying to structure it, right?Philip [00:10:44]: Right.Swyx [00:10:44]: I want you to first make factual statements, then make opinions then make bullet point summaries, have dates, have entity references have your sources for references, all these things. Anyway, so these are things that like I think people who really experiment with structural output have to really care about. But, let's, let's recurse up the stack a little bit. Before we started recording, you mentioned something really cool, which is that there's a lot of engineering that-- inference engineering that goes on when a new model provider releases a new model, right? So let's call it GLM-5.2, Kimi K3. I had previously assumed, especially if it's like, well, GLM 5 to 5.1 to GLM-5.2, like that you've supported them before. Is it that much work?What It Takes to Support a New Open ModelAli [00:11:26]: It's a lot of work.Swyx [00:11:28]: Yeah. Okay. So like, a lot of people, all you guys, right whenever a new model launch like, people rush to say like, “Oh, Hugging Face supports this, Fireworks supports this, Spacetime supports this,” and I'm like, “Yeah, of course we support it.” But what goes into that? What goes intoPhilip [00:11:40]: I think it's more than just support it too, right? It benefits the consumer a lot. Like I think it was with Kimi K2.5 or GLM-5.2 the latest, there was an inference war, right? X provider is at 90 tokens a second. The next day we're at 150. The nextSwyx [00:11:55]: I kinda kicked that off with the GLM-5.2.Swyx [00:11:58]: I wrote a Twitter article about. It got like half a million views,Ali [00:12:02]: Based on being numberSwyx [00:12:03]: YeahAli [00:12:04]: Or it's for something else.Swyx [00:12:05]: Yeah. Which,Ali [00:12:06]: Oh my GodSwyx [00:12:07]: Which then got everyone really excited about, hey, how can we, bend tracks a little bit further and,Philip [00:12:14]: There's a difference between support the model, as in I can make a token out of this model, and support a model, as in I have a production-ready API from this model.Philip [00:12:26]: Getting to the point of I can make a token out of this model is not that hard because generally the, open source inference engines, vLLM, SGLang of the world oftentimes even receive weights ahead of time, maintainers do, or the people making the model merge PRs to ensure support. So you generally can, just get it working on the standard open source stack without too much pain in most cases. The challenge is, every inference company is gonna have own proprietary stack. Some open source components, some in-house stuff. And for any arbitrary model, there's going to be some new stuff. Sometimes you get lucky, like K, two five to two six was, like, pretty similar.Quantization, Speculators, and Production ReadinessAli [00:13:16]: Yeah. It was pure continued post-trainingPhilip [00:13:18]: YeahAli [00:13:18]: If I remember correctly.Philip [00:13:19]: Even in those cases, there's still stuff you have to do. You have to redo the quantization work. You're taking the model from. Generally, these models are not released in NVFP4, and we want them to be in NVFP4 for maximum Blackwell compatibility. So we have to perform that quantization, and, calibrate the quantization to make sure that we're not causing any regression in the model's intelligence. And then we also have to train the speculator, as we've talked about. Generally, we have. We have ZDR, zero data retention on our model APIs, so we don't know exactly the traffic that people are sending us, but we know what's popular. We know that coding use cases are popular. We know that agents, agentic use cases are popular. So we can get public data sets that are representative of that traffic and train general speculators. Now, with speculators today, you need to train the speculator using the base model itself because you're getting hidden states out of the model from running inference on these specific prompts, and that is the training data you use to create the speculator. So there's that process which you need the real model weights for. And then there's of course just the process of, standing up all the infrastructure behind it, loading all this stuff, testing it. And then when there's a new model with a newer architecture, I think that, like, the DeepSeek models tend to be the most challenging as they have, like, the most novel architectural stuff going on, model after model. But every new model has something. Kimi K2 had. Oh, sorry, GLM-5.2 hadAli [00:14:53]: Sparse attention.Philip [00:14:54]: Yeah,Ali [00:14:54]: YeahPhilip [00:14:54]: the DSA.Ali [00:14:55]: Right. Which is brought from DeepSeek.Philip [00:14:57]: Yeah. AndAli [00:14:59]: So you can copy-paste then?Philip [00:15:01]: It kindAli [00:15:01]: I don't know how this works.Philip [00:15:02]: So, like we had to, like, build support for that into our runtime. And you're right, like it is really interesting the way that all of these open source labs borrow from each other. For example, like GLM-5.2 doesn't have vision. So something that, Haley, a guy on our team, if we could take a look at this, he, like, grafted the Kimi vision encoder onto GLM-5.2.Retrofitting Vision into GLM-5.2Ali [00:15:27]: We'll be training the projector.Philip [00:15:28]: Exactly. So if you think about, like, the encoder, there's the encoder, which is the part that looks at the image and turns it into latent information, and then there's the projector which likeAli [00:15:38]: You can say latent space. It's okay.Philip [00:15:41]: And then there's the projector that maps it onto, the model itself, and then there's the model weights. You don't wanna mess with the model weights because you run a chance of making the model dumber at something else for the purpose of giving it vision. So instead, Haley started with just a projector, which is only a handful of millions of parameters.Ali [00:16:02]: That would be, yeah.Philip [00:16:02]: Yeah.Ali [00:16:03]: Can you show the training one?Ali [00:16:04]: Like the way it groksPhilip [00:16:05]: YeahAli [00:16:06]: Very interesting.Philip [00:16:06]: And maybeAli [00:16:07]: That right therePhilip [00:16:07]: Maybe Ali, you should take it from here. You've got a betterAli [00:16:10]: Ooh, double the sandPhilip [00:16:11]: Understanding of this than I do.Ali [00:16:11]: Yeah. You can see, like, he. The way he trained this is really cool. At the beginning, he was training it using just like, “Here's a picture of a mountain. Can you describe what's in this mountain?” And that caused it just like the first, learning walls. Like here you can see this all we're trying to teach it is to translate the encoded. Like it's already taken the encoder from Kimi K. It's taken the image. It'Philip [00:16:31]: Yeah. FrozenAli [00:16:31]: FrozenPhilip [00:16:32]: With adapter.Ali [00:16:32]: Exactly.Philip [00:16:33]: Yeah.Ali [00:16:33]: So the brain is frozen and the eyes are frozen. It's just we're tryingPhilip [00:16:37]: AlignAli [00:16:38]: Interconnect between the eye and the brain, right? So the projector. And so you take the tokens and then he's like, “Oh, can you describe what's in this image?” And he's like, “Oh, it's a mountain,” or it's a person or it's a human, whatever the case is. But that didn't cause complete understanding. So he changed it such that every image was associated with a data set of questions. Like, does this image have a white male? Does this image have birds in the top corner? Does this image have a scientist in it? All of that stuff. And it would have to answer questions correctly. And using not just training on describing an image, but being able to answer question, another question, answer over time. Like you can see the grokking, which is like genuinely insane, that retrofitting vision into a large LLM can learn to that extent. And even for images that it doesn't perform well on, for instance, if you ask it a picture of like Stephen Hawking, “Who is this?” Maybe it doesn't get it, but it will say something like, “This is Albert Einstein.” Like it still understandsPhilip [00:17:25]: Close enoughAli [00:17:26]: That this is a scientist who is a man who has, some significant achievements, all that stuff. So that's like really cool.Philip [00:17:32]: Yeah. So, we've covered Hao Tian before, who the author of the LLaVA paper that did this, a while ago. And I think that's very foundational work for anyone who hasn't done vision work before.Ali [00:17:41]: Same with the CLIP and MetaCLIP, where you go from just captioning to building out questionsPhilip [00:17:47]: RightAli [00:17:47]: Off the image and how much better you can get performance.Philip [00:17:50]: Right. Right. Right. Yeah. But what's, what's so exciting about this is if you look at a model like this. Now, this is a little bit more of a research project. It's not. It got to 56% on MMLU Pro, I think. So not quite frontier. But if you're running this model, you haven't suffered any loss on your GLM-5.2 quality. If you don't have an image, it'll just behave exactly the way it used to. And ultimatelyAli [00:18:14]: Which in the inference code you literally do not include the other part, right?Philip [00:18:18]: Yeah. You would just skip the encoder if you don't have an image input.Ali [00:18:22]: Okay.Philip [00:18:22]: Just confirming.Philip [00:18:23]: YeahAli [00:18:23]: Does it affect a lot on the overall inference side? Like you're not adding much, you're adding a very small vision encoder. These are typically likePhilip [00:18:30]: They're super fineAli [00:18:31]: Less than a billion parameters, right?Philip [00:18:32]: Yeah. It's, - There's a little bit less standardization among vision encodersSwyx [00:18:37]: YeahPhilip [00:18:37]: So the support matrix can be a little bit, sparser. But overall, yeah, it's a pretty, it's a pretty minor component of the overall system. And ultimately what you get out of the system is all of a sudden you have Kimi Vision, GLM weights, and DeepSeek attention all in one model.Open Source Model Grafting and Franken-MergesPhilip [00:18:56]: And that's, I think, a lot of the power and beauty of open source, is that you can take all of these different components and combine them together into a system that's better than anyoneSwyx [00:19:05]: YeahPhilip [00:19:05]: Can be individually.Swyx [00:19:06]: People used to say that you would also do Franken-merges where you would take likePhilip [00:19:10]: YeahSwyx [00:19:10]: Layers from each model.Swyx [00:19:11]: Does anyone do that anymore?Ali [00:19:13]: Well, to your point previously when you were mentioning like, the work that goes into supporting a model when it first comes out, like GLM-5.2 or MiniMax M3 or whatever the case is. Sometimes you do have to like, you do have to switch out some things. Like, for instance, the MiniMax M3 head uses full attention, and with full attention you end up with this like insane bottleneck in spec dec ‘cause you're doing auto-regressive token generation for three tokens, and you're doing this like N squared over all of the tokens that are in your sequence. Your KV cache is like very large because it's not sparse, it's not top K. So we find it better to like, okay, we're gonna replace this, we're gonna replace this layer with a layer from another model that's using like GQA, for instance. And then just with the right training, you can get it to have the same acceptance rate. So it is very possible to retrofit layers from other models and very much needed. If a layer is like inefficient, the training just becomes the challenge, like how do you ensure that you train it properly? Which again to your earlier point is like the mesh between training and inference. As in like you need very good training in order to do fast inference. That's like, I feel like more and more becoming true.Swyx [00:20:21]: Yeah. Anything else on the support side when you say like get it to fully production ready?Loop Detection, Race Conditions, and Non-DeterminismPhilip [00:20:26]: Yeah. I think that there's also a question of just, we can test a model to a pretty extensive degree, but we're trying to get it out quickly and then you see a bunch of other people test it and you get interesting results. There was an issue with, GLM briefly where we had some like mode collapses where it would just output the same token over and over again for certain prompts on certain temperatures. Like once you expose an endpoint to the real world, there's going to be, so many more varieties of things given to it that you're able to, discover and patch things. So it's not just a, day zero process, it's then like for the first week, for the first month, if a model remains popular, like how do you both fix bugs and then continue to push the envelope on performance?Ali [00:21:21]: What do you mean you don't want your model outputting S?Swyx [00:21:24]: Is there loop detection on that stuff, by the way? It still happens like quite a lot, which is surprising.Ali [00:21:30]: We have like we, in our endpoint, like if a model was to output the same token like four plus times, we just cut the generation. We say like, “Oh, sorry, this-- Like try again,” or like we will reprocess the request. ‘Cause we know then, like if it, like if, yeah, it's four times the same token, it's probably collapsed.Swyx [00:21:45]: Yeah. Is there a way to opt out in case I really want that?Ali [00:21:48]: You want that?Ali [00:21:50]: I think there's a way that we have to handle it. I'm not exactly certain, but I feel like in certain models, like when they output something like you can imagine, like a table for instance, and so they want, they wanna draw like 12 dashes and 12 dashes. Yeah, I think there's a way for that to happen. I think we only do it on certain tokens. Like we exclude certain special characters.Swyx [00:22:07]: Yeah.Ali [00:22:07]: So we only do it on like certain like S is the most common almost. GLM-5.2Swyx [00:22:11]: OhAli [00:22:11]: And I think it was DSV 4 as well. Like you'd just have like looping issues where like you literallySwyx [00:22:17]: ItAli [00:22:17]: Just have like S.Swyx [00:22:18]: Yeah. Is there a special, something special about S? No, just randomlyAli [00:22:21]: It just seems to be the one token involved.Swyx [00:22:23]: Yeah. And it'Philip [00:22:24]: Is thereSwyx [00:22:24]: And it's only temperature 0Ali [00:22:27]: NoSwyx [00:22:27]: Even at other temperaturesAli [00:22:27]: Even at like 0.9 or whatever, it will still, it will still collapse.Swyx [00:22:30]: That's weird, right?Ali [00:22:30]: It's, it is an inference problem to be honest, like a software problem. Like oftentimes, the image you run will-- like NVIDIA will release an image for instance, and if we will upstream the changes from their latest TensorRT-LLM image into our stack, we'll find that it fixes it. Or oftentimes this will only happen in an inference engine that you're using like SGLang. But if you were to switch to vLLM, that isn't the case. So it seems to be like an extremely like deterministic software issue and not really a model issue. It's not like a weights problem. Like I'- we'll say like, “Oh, it's a problem with the quant. We did PTQ wrong,” right? But that isn't, that doesn't make sense because the same weights used with a different inference engine does not repeat the problem. And sometimes it's, the kernels that are being used in the backend have like these very subtle sometimes race conditions, where if you were to use this model hosted on one cluster, you will never get this problem.Swyx [00:23:19]: Oh my God.Ali [00:23:19]: But if you host it on a different cluster, you will. And the reason is the KV cache transfer from a node to node in that one cluster is using a slower interconnect than the node to node in another cluster. So that exposes the race, whereas in another cluster it doesn't. So then you end up just like, okay, this model is not gonna be hosted on this cluster. We're gonna host it on, another cluster because that cluster exposed that problem. But then it ends up with like, okay, is it the software? Is it the model weights or is it the hardware?Swyx [00:23:42]: There is a thing about this with temperature 0 still not being deterministic, right?Ali [00:23:46]: Right.Swyx [00:23:46]: Mostly because of hardware. Even at temperature 0 same model, you won't always get the same output.Swyx [00:23:52]: Even-- But I'm surprised by the race condition one because, I thought PyTorch was a graph that like guarantees that you at least, execute things in the right order.Ali [00:24:02]: Well, yeah, true. Like I'm not, I'm not saying that there is. Like well, you have things like PTL optimizations where like you can start a kernel before the end of the previous kernel, and that's like ‘cause you want to do that because there'sSwyx [00:24:12]: It's like pipeliningAli [00:24:12]: Expense. Exactly.Swyx [00:24:13]: Yeah.Ali [00:24:13]: But it'- But you don't do it cleanly. Like you overlap a little bit of the execution. No, it is very possible that the kernel itself, like that one block that is supposed to be running in this instance of time, that kernel itself has a race condition. For instance, like a missing barrier. Like often if you're designing a kernel and you want it to make it to be very fast, if you don't test it extensively, you'll, you'll have certain threads access data points from registers before they've been written to by other threadsSwyx [00:24:36]: YeahAli [00:24:36]: For example, because like your barrier is wrong or your synchronization was wrong. But yeah, like the testing itself is very difficult in those like, andSwyx [00:24:42]: And there's no like borrow checkerAli [00:24:45]: What does that mean?Swyx [00:24:46]: Like Rust. Like the. If you're trying to have like memory safety It sounds like a comparable problem.Ali [00:24:52]: Well, yes, but you're working in CUDA, right, NVIDIA GPUs. Like- You just need a higher level language like modular Maybe that's what modular is supposed to do. I don't know.Quantization Quality and Vendor FidelityVibhu [00:25:00]: How do you see keeping quality of the model? So you talked about all these steps of, okay, you gotta do quantization, train your own speculative decoderAli [00:25:07]: RightVibhu [00:25:07]: Run on different hardware. Looking at other model providers, okay, you kicked off a inference speed race on the consumer end. What goes into keeping quality the same across them, right? Sure, you can run benchmarksAli [00:25:22]: YeahVibhu [00:25:22]: But, like, how do you determine how much quantization are there standards? What goes intoPhilip [00:25:27]: There's a few things on quality. Most inference optimizations are lossless. KV caching, for example. You are just recomputing or preventing recomputing the same values. Speculation, of course, if a draft token is wrong, it gets rejected. The main lossy optimization is quantization. And that really comes down to, number one, data format, number two, which parts of the model you choose to quantize, which layers, and number three, like doing a lot of calibration on the quantized weights, to ensure that you're preserving all the outliers. There's other tricks that you can do, though. A big one is long context, ‘cause one thing you asked at, right at the beginning is, “Oh, what's gonna happen if I send a 200,000 token request in?” So with a long input sequence, you need to, store a lot more information. You need to process a lot more tokens. And so even if a model has a context of a certain length, you might, as an inference provider, choose to build an API with a shorter context length, and of course a full length one as well. Because if someone doesn't need the full million token context, for example, you can get them better performance. I don't know if that's exactly like quality of the model. The way that I think about quality is to what degree are we faithfully serving the original model? If you think of a golden implementation of a model that performs exactly the way the model is designed to perform, I think of quality as how close are we getting to that, 100% fidelity of the model.Philip [00:27:13]: You can also, of course, think about quality from the training side and how do you push yourself past 100%. But when I think about purely inference optimizations, it's getting faster while staying as close to that 100% fidelity mark as possible. And certainly our standard internally is that, like you should not be able to tell the difference between our API and a, official API. I think Kimi in particular does a good job of vendor benchmarking hereAli [00:27:41]: YesPhilip [00:27:41]: Where they haveAli [00:27:42]: They released an actual vendor benchmark.Philip [00:27:43]: Exactly, yeah.Ali [00:27:44]: ‘Cause they accused, some people, Amazon? There was some provider that was not doing very well on Kimi's benchmark.Philip [00:27:50]: Yeah.Philip [00:27:51]: So, with Reflect we probablyVibhu [00:27:52]: This was a long time ago, right?Philip [00:27:54]: No.Ali [00:27:54]: Yeah, like threeVibhu [00:27:55]: They alsoAli [00:27:55]: Four, five months agoVibhu [00:27:57]: This also happened with, I don't remember which model, but they pulled out quite a few, and then they started a whole chart about this. It might have beenPhilip [00:28:03]: Kimi Vendor Verifier.Ali [00:28:04]: Yeah.Philip [00:28:05]: Yeah.Ali [00:28:05]: Yeah, ‘cause you, ‘cause you'd be pissed, right? Like if you'Philip [00:28:07]: Yeah.Ali [00:28:07]: If like if I'm a consumer and I'm using like Amazon's endpoint for instance, and I've used Kimi and I'm like, “Oh my God, like this is bad,” I'm not gonna say, “Oh, Amazon quantized the model in a bad way.” I'm gonna say, “Oh, Kimi sucks.” Right?Philip [00:28:17]: Yeah.Ali [00:28:17]: So it seems like that makes sense.Philip [00:28:19]: Yeah, they care. They care.Vibhu [00:28:21]: Justifiably.Ali [00:28:21]: Yeah, justifiably.Vibhu [00:28:22]: This is probably a stupid question, but just checking, has anything improved from main quantization?Philip [00:28:28]: Yeah.Vibhu [00:28:28]: Like, is quantization always strictly worse?Ali [00:28:30]: Well technicallyVibhu [00:28:32]: NoAli [00:28:32]: It's a lossy. QuantizationPhilip [00:28:33]: YeahAli [00:28:33]: Is a lossy, it's a lossy implementation.Philip [00:28:36]: Speed improvesVibhu [00:28:36]: Speed improves.Ali [00:28:37]: It the number, likeVibhu [00:28:38]: No, I' always look for inverse scaling laws.Philip [00:28:40]: Yeah.Ali [00:28:40]: Yeah.Vibhu [00:28:40]: This is something I learned from Noam Brown, where like things that normally act in one direction sometimes do.Philip [00:28:45]: Well, technically when you run a benchmark, because these models are deterministic, sometimes your,Ali [00:28:52]: YeahPhilip [00:28:52]: NVFP4 quant is like, two basis points higher than yourAli [00:28:56]: No, it's noise. It's noise.Philip [00:28:57]: Yeah, exactly. I'm like, yeah, it's, it's within. That's why I always say within margin of error.Philip [00:29:01]: And I stopped saying that because everyone assumes that what is, well, within some margin of error, we're barely inside of that to the worst, so we're saying. But yeah, sometimes it's just like, gives you a higher output score. But like Ali said, that's noise. To my knowledge, you're not necessarily making the results better. You're just trying to, again, like keep your fidelity as close to 100% to the original model.Layer Selection, KL Divergence, and Better QuantizationAli [00:29:27]: There is, to your point, research that we did on MP. I don't know if you are able to pullPhilip [00:29:31]: YeahAli [00:29:32]: A tweet we did. One of our research interns, Joshua, I think it's a tweet on how we have 20% better quantized GLM-5.2 than NVIDIA. Essentially what we found throughout like this month research is, okay, quantization is a lossy. It's. You're compressing the data from, occupying 16 bits to occupying, four bits, for instance. And so you're losing some information, and you're trying to minimize that. And so when I say that I'm gonna quantize the model, my job becomes how do I find the layers that I can quantize, and how to find the layers to not. For instance, with image models, I don't quantize modulation layers, and I don't quantize out projections because those two are. Like out projection is what you see as the user. Modulation is what the model sees or understands. Right, exactly. And so to his paper, do you have the. It doesn't have the. Yeah. It's a long paper. I don't know if I can findVibhu [00:30:25]: If there's a part to search or it's probably in the thread.Ali [00:30:28]: It's probably in the thread.Vibhu [00:30:29]: Yeah.Ali [00:30:29]: But the long and the short is it is very possible that quantizing more of the model makes the results. Like if I have a model that I quantize layers one, five, and 10, and another model where I only quantize layers one and It is possible that the model in which I quantized more information is going to perform better because the quantization errors have canceled out. And so what Joshua showed in his mathematical proof where he had like a verifier in, is that you can predict which layers are going to have quantization errors that will cancel out with each other, and you choose to quantize those layers. And so the result of doing this mathematical quantization is you end up with a model that's 20% more quantized than another provider, so you get 20% more throughput of it because there's more layers than running an NVFP4, and your quality is better than that other quant because the layers that you chose to quantize have their errors cancel out, like one layer skewed to the right one layer skewed to the left, one layer skewed to the right. Your final logits distribution is more similar to the original distribution of the model, so you have better fidelity. And so the way we proved this was with KL divergence. So instead of just scoring on the benchmarks, we scored the KL divergence between the logit distribution of the quantized model and the logit distribution of the original full precision model, and we showed that with this technique we get. If your probability distribution on the logits which token it wants to select is more of the same as the original model, you're probably gonna end up staying true to the original model. So yeah, so it seems like previously before this, it seemed like the industry was, well, the more you quantize, the worse it's gonna be, ‘cause the more loss you introduce. That's not exactly, not necessarily true. So yeah, doesn't improve it, but can cancel out.Philip [00:31:57]: I think it might be this, but reminds me a good bit about pruning where you can prune off certain layers.Philip [00:32:03]: But very interesting. Didn't know this was a whole paper you guys put out.Ali [00:32:06]: It's. Fun fact, it was originally 72 pages, this paper, and then we decidedPhilip [00:32:11]: WowAli [00:32:11]: We can't tell. We couldn't release it. So it's now 45.Swyx [00:32:15]: Still 39 pages, so very substantive. We talked about evals and all these things and, like what's possible in terms of speedup? Like it's like probably like the numberInference Speedups and BenchmarkingSwyx [00:32:25]: Thing that people do wanna care about, and it's something that you wrote about in your post. Like official API is 70 tokens per second, and you push it up to 90. Is that like a normal thing?Philip [00:32:36]: So what's cool about working in inference, the reason that I think inference is going to be a useful place to do engineering for a long time, is that if you look at highly optimized domains like, say, finance, if you're in finance, you measure how much better you got in basis points. It's like, “Oh, I got five basis points better, like twentieth of 1% better,” that's huge news because everything is so optimized. When we publish optimizations, it's 20%, it's 100% it's 200%. So there's still probably like a lot further to go, honestly. Like you'll, you'll know that inference is pretty much solved when researchers start publishing about how they got 1% faster at something.Swyx [00:33:19]: Which by the way, because I am from the finance background, in the ‘70s, that was the margin at the time. When you did quantitative finance research, you would findAli [00:33:27]: And like 20%, tens of percent.Swyx [00:33:29]: That's. Yes.Philip [00:33:29]: Yeah.Swyx [00:33:30]: And now it'Philip [00:33:31]: Tiny fractionsSwyx [00:33:32]: For those people interested, look up Andrew Lo's paper. He had a really interesting illustration of quant, stat arb, distribution, narrowing down from like those kinds of 20% differences in the ‘70s, down to nothing today, which is very cool.Philip [00:33:48]: Exactly, and we're at the beginning of the same type of thing. Now benchmarking is hard. I think anyone will tell you that, and benchmarking provider speeds is hard because there's so many variables that go into it. What hardware are you using? How much load do you have on the system? What's the exact nature of the prompts and input and output sequence lengths? All that stuff. But overall, when you start stacking these improvements, you're looking at multiples. You can look at it. The most common form, of course, is TPS, tokens per second, which is bad naming by us in the industry, ‘cause there's two tokens per second. There's tokens per second, the throughput number, and the latency number.Ali [00:34:31]: TTMT, yeah.Philip [00:34:32]: Like total tokens per second out of the, out of the GPU as a throughput number. Most people only care about tokens per second as the latency number, which we should call ITL, intertoken latency, but we don't.Philip [00:34:44]: Anyway, so you can imagine a standard API without many optimizations for a 1 trillion parameter model operating somewhere in the 30 to 50 tokens per second range for reasonable traffic profile. And we generally see the goal of, pushing to 10X that. But, not necessarily day zero, but by stacking enough optimizations, if you have, say like four optimizations, each of which doubles performance. Or sorry, three optimizations, each of which doubles performance, then you stack that up, that's an 8X gain. That's the order of magnitude that we're working with in this space. We're trying to make things substantially faster, not just go from like 70 to 90.Swyx [00:35:38]: Are you saying you've. You have done that?Philip [00:35:40]: So let's say you have as a reasonable baseline, 30 or 40 tokens per second. You can achieve 10X that. So like on GLM-5.2, if you run it unquantized, perhaps on H100s even, and you're just using an off-the-shelf inference engine with no particular optimizations, no speculator, nothing extra around like KV routing, no disaggregation, you're, you're probably, yeah, looking at that like 30 to 40. You think that's like a reasonable baseline?Swyx [00:36:12]: Right. Right.Philip [00:36:12]: To get to something like 10X, there's a lot of trade-offs that you're making. If we're running at more like a 300, 400 tokens per second range, you are using the best hardware possible. You have a optimized speculator. You have done all of your quantization work. You are Seeing a pretty high cache hit rate. You are running with a reasonably small batch size and a parallelism configuration that is tuned for latency versus throughput, but it is possible. So the spreads that you see if you, like, go on artificial analysis or you go on OpenRouter and you look at, the worst provider to the best provider, oftentimes can hit that range. 10X is of course very aggressive. It's oftentimes maybe more of a four to six times improvement. But that's the performance that makes us really excited, is when we can get these huge gains, not just go from 70 to 90 tokens.Stacking Optimizations: NVFP4, Speculation, and DisaggregationAli [00:37:19]: It's also, like, hardware dependent. Like, ifPhilip [00:37:20]: YeahAli [00:37:20]: If you have a thing where you're serving it on just, like, a node of H100s and then you throw, like, you shard the model across, like, four nodes of B200s. Like, you can definitely increase the speed with just throwing more hardware at it. Like, normalizing for the same exact hardware and the same number of GPUs.Philip [00:37:35]: Yeah. Then you're looking at, like, a two to 4X improvementAli [00:37:38]: Right. RightPhilip [00:37:38]: Depending on the inference optimizations. So yeah, it's. Some of it's, what's the call, and some of it's who's the driver.Vibhu [00:37:46]: If you break down the two to 4X, say the example is run GLM-5.2Ali [00:37:51]: YeahVibhu [00:37:51]: On B200sAli [00:37:53]: YeahVibhu [00:37:53]: Single node, right? What's, like, the cost trade-off for effort to get, like, the last bit of juice out versus what should people just think of, right?Ali [00:38:01]: Spectre quantization. Yeah.Vibhu [00:38:03]: Spectre quantization.Ali [00:38:04]: That's, that's, that's like 95%. LikeVibhu [00:38:06]: And how far does that get you? And how easy is that for the average person to do? So say right I wanna throw the weights of GLM-5.2 on a node of B200s, how easy is it to find speculative decoder- decoder model or already quantized model? How much work goes into it?Philip [00:38:23]: If you're doing it up front, it's quite a lot of work. If you're doing it today, there's going to be people who have published things that you can just, you can just grab some NVFP4 weights. You can grab a speculator. Yeah, if we're thinking about, like, what are the 2Xs we're stacking, going from, BF16 to NVFP4 is, it's not quite a 2X, right? It's like. I think it's about, like, 30 to 40%, from 16 to 8, and then another 30 to 40% multiplied from, 8 to 4. So that doesn't quite get you a 2X, but, like, roughly a 2X. Speculator, roughly a 2X. Disagg on top of that if you're able to get enough hardware and put enough traffic through it, another roughly a 2X. And then you add in some, double-digit percent increase from having just a better runtime with, the latest kernels and stuff behind it. And that's how it stacks up.Ali [00:39:21]: YeahPhilip [00:39:21]: So building each of those, like, building the, quantized weights is, for someone who really knows what they're doing, hours to days of work. Building the speculator, again, like, hours to days of work. And the, disagg setup, hours to days. Well okay, but like once you haveAli [00:39:39]: Once set up. Once set up. YeahPhilip [00:39:40]: Yeah, getting disagg working for the first time, I'm saying, of course, is very difficult.Philip [00:39:44]: The marginal implementationAli [00:39:48]: Like, if you're just grabbing, like if you are a person, like just a normal consumer who has access to, like, a node of B200s and you're wondering, “How can I just host it myself?” You don't need to quantize the model yourself. There's always gonna be, like, an open source quantized checkpoint. NVIDIA's gonna push one out if no one else does. You. Usually, the providers will have their own spec dec that they've trained as well. You don't need to train your own spec dec. You can just use that as well.Philip [00:40:09]: Yeah. Like, GLM-5.2 has its own MTP.Ali [00:40:13]: Right. Right.Vibhu [00:40:14]: What's multi token prediction?Philip [00:40:15]: Yes.Ali [00:40:16]: I'm justVibhu [00:40:16]: Can you explain that?Ali [00:40:16]: I'm just an expert.Ali [00:40:18]: I can do it for you in case I get it wrong?Vibhu [00:40:20]: No.Vibhu [00:40:21]: Yeah, you should correct if we're wrong, but their multi-token prediction can be used for self-speculative decoding.Ali [00:40:27]: I'm not sure. I'm not gonna correct that.Vibhu [00:40:28]: Okay. I'm semi-confident in thatAli [00:40:30]: Okay. YeahVibhu [00:40:30]: But someone can check. But it's useful to paint the story of, okay, not just the average person, but say a company wants to switch from serverless inference I wanna throw this up on. I wanna rent some GPUs, throw it up. These are the steps you take to do significantly faster than just put it behind vLLM.Ali [00:40:48]: Right.Vibhu [00:40:49]: I was waiting for a mention of Dynamo.Vibhu [00:40:51]: I feel like, that's supposed to be the baseline that you measure against.Dynamo, KV Routing, and Disaggregation ToolkitsPhilip [00:40:55]: I would think of Dynamo as less of a box system and more of a toolkit for building with. So when we talk about doing aware routing, when we talk about doing KV offloading, when we talk about doing, PD disaggregation, Dynamo fundamentally is. By the way, Dynamo is an open source library from NVIDIA.Ali [00:41:17]: We've done a pod with KylePhilip [00:41:18]: OkayAli [00:41:19]: Kyle Cranin.Philip [00:41:19]: Cool. So then your listeners know then that it supports all the different inference frameworks. And it is multi hardware, which is interesting.Ali [00:41:28]: But it's just a router, it's not like an optimizer layer.Philip [00:41:30]: Yeah. All it does, like, what Dynamo is good at, it is a library for moving information around your cluster, around your hardware. So if you have, KV cache on one place and you need it to be somewhere else, Dynamo coordinates NIXL for you to move that around.Philip [00:41:49]: That doesn't mean that, like, out of the box, you just say, “Pip install Dynamo,” and then you get, like, a massive performance speed up. It's more of a developer toolkit.Ali [00:42:01]: Yeah. I would have said it would. It comes with a set of defaults that you can then swap out.Philip [00:42:06]: It does. If the industry at large, I think, was, like, rolling out all of these deployments, standard, then I think it would be, like, a credible baseline. But, we've got to, we've got to benchmark against, like, what we're seeing in the wild.Speculative Decoding Methods: Medusa, EAGLE, n-Gram, and Spec-SpecVibhu [00:42:23]: I did wanna talk a little bit more about PD disagg, because that is probably, like, number three after quantized and speculative decoding. In your book though, I was just gonna pull out the book.Philip [00:42:31]: Yeah.Vibhu [00:42:32]: Like section 522 on Medusa, 523 on EAGLEPhilip [00:42:35]: YeahVibhu [00:42:36]: 524 on gram.Philip [00:42:37]: It's 55, would be disaggregationAli [00:42:42]: Yeah. Well, no, I just wanted to dwell a little bitPhilip [00:42:44]: YeahAli [00:42:44]: The other. Like, so what do you choose to include? What do you choose to not to include? Because there was all these other techniques.Philip [00:42:51]: Yeah.Ali [00:42:51]: Are these still relevant? Because I think they came out, like, a year and a half ago maybe.Vibhu [00:42:55]: Medusa is quite old.Philip [00:42:56]: Yeah, Medusa's old.Ali [00:42:58]: It was old.Vibhu [00:42:58]: But is it in the book as a good, here'sPhilip [00:43:01]: BaselineVibhu [00:43:01]: Baseline vanilla understand it?Philip [00:43:02]: Like you should know this.Vibhu [00:43:03]: Like I read the paper, I'm like, “ it makes so much sense.”Philip [00:43:05]: Yeah.Philip [00:43:05]: So with the book, I had a couple goals. One was to give people just a working vocabulary for the space as a whole, and the other was to give them some intuition about how each of these techniques works. As I mentioned in my AI Engineer talk, which is the first public addendum to this, the speculation space has moved much faster than everything else. So yeah, even at the time that I wrote the book Medusa, I very much included as a way for people to understand how the space evolved rather than what the most modern technique is. And now of course, there's DFlash, dSpark. There's, there's newer techniques even than EAGLE, although EAGLE is still very commonly used.Ali [00:43:51]: SpecSpecta.Philip [00:43:52]: Yes. Speculative decoding.Vibhu [00:43:54]: What canAli [00:43:56]: Oh, it's a paper by Tri Dao and it's like, it's doing speculative decodingVibhu [00:44:00]: HuhAli [00:44:01]: For the speculative decoder.Philip [00:44:02]: Oh, in spec- oh my God.Ali [00:44:02]: It's literally just an another. It's like, yeah, that's the most simple way to explain it, and it seems like he got trivial speed ups there. But it seems that the complexity with training, it's almost like in our mind at least, it's almost as complex as training GANs. Like it's like a very delicate balance and oftentimes you, it's just but yeah, it's literally speculative decoding on speculative decoding.Vibhu [00:44:21]: Speculative.Ali [00:44:22]: Yeah. We saw this paper.Vibhu [00:44:24]: It's interesting, right?Ali [00:44:24]: Yeah.Vibhu [00:44:24]: I wouldn't even expect it to be very particular to train, I wouldAli [00:44:29]: Right.Vibhu [00:44:29]: The naive part of me is like, okay, train speculative decoder.Ali [00:44:32]: But like, and it makes sense, like the whole idea of speculative decoding is you. It's like, it's like almost like the iPhone auto predict version but for a normal model, right? Like you're just, you're just, generating three tokens and you're like, okay, I'll do prefill on them. And so you save those three turns for your original model. Now your speculative decoder is doing three turns of auto regression, so why not just have an even smaller model?Ali [00:44:53]: The other question there is what are the size of speculators? So say forPhilip [00:44:58]: Right. It's like a billion parameters.Ali [00:45:01]: Like for MiniMax, it's. Yeah. It's like one layer. It's like one 60th of the original model usually.Philip [00:45:06]: Yeah. I think we should do a paper when we get back to the office.Philip [00:45:10]: SpeculativeAli [00:45:11]: SpeculativePhilip [00:45:11]: Decoding.Ali [00:45:13]: No, it's, it does seem like how, when do you stop? But then it also seems like if you're able to train spec-spec decode for instance, right? Like if you're able to have a small model that is accurately predicts what the intermediate speculator is gonna predict, that is able to predict what the original target model's gonna predict, then why not just use that smallest model directly, right?Vibhu [00:45:34]: Yeah. This isAli [00:45:35]: Like it seems likeVibhu [00:45:35]: Adjacent to the routing problem.Ali [00:45:36]: Right.Vibhu [00:45:36]: Yeah.Ali [00:45:36]: Right.Philip [00:45:37]: The thing with speculators is one of the practical constraints on using them is that you do have to run a small model on the same hardware that you're running the big model on. There is a orchestration and resource competition problem inherent in that, and that is one of the constraints on speculation in general, is that draft tokens cost resources to create and cost software complexity to manage. And so if you have like infinitely recursive speculators, you add in quite a bit of that complexity on the actual implementation within the inference engine as well, not just in the training process.Vibhu [00:46:17]: I was gonna say, I would wonder if you could do similar, like distillation and pruning of, it's the same thing, it's just a model. Can we not just distill a lot of the weights, quantize the speculator, out of my domain? The question that also comes up is, this is all for big server workloads, right? How much of this applies to, say I have this MacBook, I wanna run Gemma really efficiently. Similar problems, not the same?Local AI vs. Data Center InferencePhilip [00:46:45]: Pretty different. I talked to Selo, about this on his podcast a couple weeks ago. The difference between inference engineering for the data center and for production workloads versus inference engineering for local AI, is that we start with fundamentally like different constraints and different goals. With local AI, it's how do I fit this model onto my hardware and then make it less dumb? And with data center influence, it's how do I load this model and then make it less slow? And we care about less dumb, and they care about less slow. But the local AI inference engineering ecosystem, I think has a lot for us to learn from in the data center space. They are experts in various forms of quantization, including dynamic quantization that we just don't touch, in the pruning, in the distillation, in the, layer removal. There'Ali [00:47:42]: Layer removal matters less.Philip [00:47:43]: Yeah. There'Ali [00:47:44]: No one loves pruning really.Philip [00:47:45]: Yeah. Well, but the, but they doVibhu [00:47:46]: Which is surprising, right? But that's, that's a whole different thingPhilip [00:47:48]: Just to fit something on the laptop.Ali [00:47:50]: Right.Philip [00:47:50]: So yeah, it's a, it's an interesting, it's an interesting space. Not necessarily that like their techniques make sense for us to do in the data center, because we have different resources and different goals, but more that the process as well as the openness of that field is something to, admire.Ali [00:48:12]: Yeah. Like to your point, like, certain optimizations that would. Like for instance, Turbo Quantum Sharper, like it made such huge hype on that and we did like a whole deep dive on Twitter and like said, what is it? How does it work? Why is it good or not? And it took off and it was implemented on local devices because your memory bandwidth is so slow on like a MacBook, for instance. But try putting the same thing on like an NVIDIA GPU on a B200 Turbo quant would not be. Like, it would not be used. Like, NVIDIA - Like, NVIDIA made it clear that this is not a good optimization, and we've seen it firsthand where the overhead of doing dequantization, quantization of, in the kernel itself with turbo quant kernel, each end is much slower than the time that you save from doing the bandwidth. ‘Cause on the B200s, you have like 3.5 terabytes per second. You don't need decrease the storage that much. You don't need to do, FP4 KV cache. You don't need to use a requant. There's, there's, there's better optimizations to be made. But on Edge devices, it's extremely important, it's extremely useful. So, seems to be, like, different optimizations there, but then they're all uniquely combined with like all you wanna quantize the model, you wanna do speculative decoding, like certain common prefixes with bothPhilip [00:49:18]: Principles.Ali [00:49:19]: Yeah, exactly. Exactly. Exactly.Philip [00:49:20]: They also do a lot of work on, model parallelism, especially over, heterogeneous topology, where you have, some sparks and they are wired together with, Ethernet, DGX sparks.Ali [00:49:35]: Yeah, this is the Exo Labs guys.Philip [00:49:36]: Yeah. You have, a nu

-Normally, Apple has no problems making enough MacBooks to feed demand. The MacBook Air shortage is linked in part to the ongoing "Ramageddon" memory crunch caused by the insatiable demand from new AI data centers. -Based on the judge's order, he was not persuaded by xAI's actions that the law's enforcement would cause immediate harm. He noted that xAI filed the lawsuit on July 29, 2026, almost three months after the law was signed and merely three days before it was set to take effect. -Seven water and wastewater utility companies have already been hit by cyberattacks since July 27, 2026, which led to degraded water operations. Learn more about your ad choices. Visit podcastchoices.com/adchoices

The Fork In Your Ear Ep#218 Jimonthy Crickets! - Podcast Show Notes & Summery 8-1-26 Quick Summary Tim and Nate open with the usual technical chaos—Nate's ancient camera older than his kids, USB hub murders, Windows being Windows, and a new meshy chair—before Tim scrapes himself off the floor with minimal coffee. The big gaming bombshell is Nintendo snagging a FromSoftware exclusive (Dustbloods) led by Miyazaki himself: Bloodborne-vibes, industrial revolution setting, up to eight-player PvE, and Tim is currently in the closed NDA network test. Microsoft brings four original Xbox classics to PC (Fusion Frenzy, Crimson Skies, Conker's Live and Reloaded), sparking preservation vs. "just sell it again" debate. Nate digs into the new Halo campaign on PC while Tim's wrist is still sidelining him. Entertainment covers Spider-Man: Brand New Day (Tim is all-in), American Psycho after a four-and-a-half-hour serial-killer museum visit, and the couple literally dodging the Bite of Seattle shooting by having a fight and going home. Life wraps with raccoon babies, the local "Jimmathy" meme, and Nate's wife getting back on two wheels with a nearly-new 2025 Triumph Speed Twin 900. Classic Fork energy from start to duck-call finish. Detailed Show Notes

The Apple Upgrade program has been revealed, so it's time to discuss what leasing your Apple products actually means. Plus, your hosts discuss Apple's plan to combat the chip shortage on the AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:Claude by Anthropic: Check out Claude and Claude Pro at Claude.ai/appleinsiderLinks from the Show:Apple Upgrade: What you're going to pay per month for iPhone, Mac, or iPadApple Upgrade goes live as a new installment plan for iPhone, iPad, Mac, & Apple WatchApple Upgrade will be great for Apple, might not be good for the buyerMicron urges White House to reject Apple's blacklist memory planUS senators urge Apple to abandon plans for Chinese-made chipsPublic distrust in smart glasses will be a challenge for Apple GlassApple's Home Hub & HomePod update may arrive as soon as OctoberTernus says he plans to build on the success of Apple TVNo major reset is coming as John Ternus prepares to take over AppleGame development diary: Launched, but far from finishedSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Apple Upgrade (19:34) - Apple and China (37:45) - Apple Glasses and privacy (46:23) - Apple Home Hub (53:21) - Apple TV's future ★ Support this podcast on Patreon ★

Apple just replaced its iPhone Upgrade Program with a device leasing model run by Klarna, and you might end up paying more for devices you never actually own. Find out what happens when your monthly payments end in this episode covering Apple Upgrade! Apple launches Apple Upgrade device lease, replacing iPhone Upgrade Program Klarna partnership explained: application, credit check, and payment limitations Lease requirements: activation, payments, trade-ins, and pickup rules Device condition, damage, and loss policies under Apple Upgrade Canceling early, returns, and keeping financed accessories End-of-term options: upgrade, return, or buy device outright Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord. Sponsor: threatlocker.com/twit

Apple just replaced its iPhone Upgrade Program with a device leasing model run by Klarna, and you might end up paying more for devices you never actually own. Find out what happens when your monthly payments end in this episode covering Apple Upgrade! Apple launches Apple Upgrade device lease, replacing iPhone Upgrade Program Klarna partnership explained: application, credit check, and payment limitations Lease requirements: activation, payments, trade-ins, and pickup rules Device condition, damage, and loss policies under Apple Upgrade Canceling early, returns, and keeping financed accessories End-of-term options: upgrade, return, or buy device outright Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord. Sponsor: threatlocker.com/twit

Apple just replaced its iPhone Upgrade Program with a device leasing model run by Klarna, and you might end up paying more for devices you never actually own. Find out what happens when your monthly payments end in this episode covering Apple Upgrade! Apple launches Apple Upgrade device lease, replacing iPhone Upgrade Program Klarna partnership explained: application, credit check, and payment limitations Lease requirements: activation, payments, trade-ins, and pickup rules Device condition, damage, and loss policies under Apple Upgrade Canceling early, returns, and keeping financed accessories End-of-term options: upgrade, return, or buy device outright Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord. Sponsor: threatlocker.com/twit

Watch The FULL EPISODE on Patreon!https://www.patreon.com/c/THEUNCUTPODCASTImportant: To avoid paying extra fees, please don't subscribe through the iOS App Store.Instead, subscribe via a PC, MacBook, or laptop using your web browser. This way, you'll avoid the additional charges applied through the App Store.Send us your dilemma here: https://uncutpodcast.komi.io.Follow us on our personal Instagram accounts:Beatrice - https://www.instagram.com/beatriceakn/Tammy - https://www.instagram.com/tammymontero/Sharon - https://www.instagram.com/sharonodu/ Hosted on Acast. See acast.com/privacy for more information.

This episode is brought to us by Back Market, the world's largest pure-play premium refurbished tech marketplace. Back Market's mission is to make refurbished tech the mainstream choice, by extending the life of devices, fighting electronic waste, and challenging today's throwaway culture.On today's show we sat down with Lauren Benton, General Manager of Back Market in the US, to talk about the real cost of family tech - and how to bring it down. To learn more, visit BackMarket.comApple just raised prices on several products (up to $300 more on certain MacBooks and iPads), and analysts expect the iPhone to see a hike this fall too. So the timing couldn't be better to rethink how we're spending on tech.Here's what we learn:Why "refurbished" doesn't mean what most people think it means, and how devices are quality-checked before they're ever resold.Where the real savings are: specific numbers on phones, laptops, tablets, and watches.The hidden costs in your tech budget that have nothing to do with the device itself.How to know when it's worth trading in an old device instead of letting it collect dust in a drawer.What to prioritize (and skip) when Back-to-School shopping collides with these new Apple prices.Learn more about Farnoosh's upcoming literary workshop Book to Brand. Early bird registration is now open! Hosted on Acast. See acast.com/privacy for more information.

On this week's episode we talk about:TikTok Creator mike_904_ says a single mother chose her son over himA 14-year-old girl escaped a 10-day captivityA Illinois State Police Trooper was investigated after a local restaurant executive, whom he arrested for DUI, used Apple's "Find My" feature to locate his missing MacBook inside the trooper's home11-Year-Old Boy with Autism Takes Uber to Airport Without Parents' KnowledgeTwitter: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://twitter.com/PnLJudgementals⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠TikTok: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.tiktok.com/@pnljudgementals⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Facebook: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.facebook.com/PnLJudgementals⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Instagram: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠https://www.instagram.com/the__judgementals⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠Email: ⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠⁠pnljudgementals@gmail.com⁠Music: Bread Crumbs - Successful

The Hide My Email failure is real but nowhere near as significant as it's being portrayed, plus Apple is raising so many prices yet looking at ways to make that palatable, plus Foxconn has begun its annual recruitment drive as it begins producing millions of new iPhones, all on the AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:MasterClass: Get 15% off annual memberships at MasterClass.comLinks from the Show:Seven years after Apple Card, Samsung leaps into fintech with its own credit cardHide My Email class action lawsuit seeks payout without evidence of any attacksIt's easy to find a real email address behind Hide My Email, but it doesn't really matterHide My Email flaw still worked two weeks after Apple's claimed fixA fake Hide My Email header can expose the address behind your Apple Account -- a different thingApple Music gets first U.S. price hike in four yearsIs Apple One worth it in summer 2026?AAPL capitalization squeaks past NVDA, Apple becomes world's most valuable companyApple's iPhone Upgrade Program was great while it lastedApple Upgrades will let users lease iPhones and Macs with KlarnaIt has begun: Foxconn amassing army of workers for iPhone 18 Pro assemblyUpgraded Mac Mini, Mac Studio, OLED iMac readied, release date hazyApple preps MacBook Pro, MacBook Neo refresh in bigger AI pushApple's $634M payment to Masimo now set in stone after Judge tosses appealCaleb Hammer on YouTubeSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Intro (02:49) - HIde My Email (16:07) - Apple Services and price rises (25:56) - Apple Upgrade (42:09) - Masimo ★ Support this podcast on Patreon ★

iOS 26 conveniently puts you in charge of your iPhone home screen with deep design tools hidden behind a single long press. Discover how Apple's new Liquid Glass and tinted themes finally let your wallpaper and icons reflect your style. Jiggle mode and edit tools enable deep icon appearance tweaks Exploring Light, Dark, Clear, and Tinted home screen styles Using tints and saturation sliders for unique color coordination Rearranging app icons to highlight wallpaper and key visuals Adding, editing, and customizing widgets on the home screen Hiding, deselecting, or deleting home screen pages safely Removing unused home screen apps without uninstalling them Switching between app icons and widgets for streamlined access Face ID and privacy tweaks for individual apps on the home screen Host: Mikah Sargent Download or subscribe to Hands-On Apple at https://twit.tv/shows/hands-on-apple Want access to the ad-free audio and video and exclusive features? Become a member of Club TWiT today! https://twit.tv/clubtwit Club TWiT members can discuss this episode and leave feedback in the Club TWiT Discord.

Apple is raising its subscription prices for both Apple Music & select Apple One bundles. The OpenAI & Apple legal battle is escalating. And are Siri's AI features finally coming to China? Apple raises prices for Apple Music and Apple One subscriptions. Is Apple One worth it for you after recent price hikes? iCloud $32.8B CSAM lawsuit dismissed, Apple protected under Section 230 laws. Apple in talks to settle DOJ antitrust lawsuit, per report. Class action accuses Apple of misleading users about Hide My Email's privacy protections. OpenAI and Apple legal battle escalates, poached employees warned about data deletion. Apple Intelligence finally on the road to release in China. PrismML releases Bonsai 27B, claiming first major AI model of its size fit for iPhone. Apple's New Speech API vs Whisper: The First Real Benchmark. Apple testing 'Live Notes' AI system to record Genius Bar sessions. Apple's scrapped Mac Pro plans reportedly included a new Intel model. visionOS 27 Beta: 90-Hertz-Freeze bei 24p-Quellen und neue Controller-Plattform. A.I. 'Vibecoded' apps are flooding Apple's app store. Investigation reveals dozens of disguised gambling apps on the App Store in Brazil. San Francisco demands Apple and Google delete AI 'nudify' apps from App Stores. Picks of the Week Jason's Pick: Breville One-Touch Tea Maker Christina's Pick: Nativ Andy's Picks: David Bowie NASA Bolt Sticker & dbrand MacBook skin Hosts: Leo Laporte, Andy Ihnatko, Jason Snell, and Christina Warren Download or subscribe to MacBreak Weekly at https://twit.tv/shows/macbreak-weekly. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: ethos.com/macbreak cirasync.com/MacBreak

The Pirate Street Journal takes a sharp look at business through the category design lens, and this episode delivers three stories that reveal how the decisions made today will define economic winners and losers for decades. From data center legislation in New York to Apple raising prices and a Costco cashier becoming a millionaire, each story points to the same underlying truth: the category you choose matters more than almost anything else. Whether you are a governor, a tech executive, or an hourly worker, picking the right side of the S-curve is everything. This is just one of the topics that Pirates Christopher Lochhead, Eddie Yoon and Bri Clark discuss on this episode of Pirate Street Journal. Each week, the Category Pirates pick three headlines worth paying attention to and break down the category underneath. You're listening to Christopher Lochhead: Follow Your Different. We are the real dialogue podcast for people with a different mind. So get your mind in a different place, and hey ho, let's go.   New York Said No to Data Centers and It Will Pay the Price On July 10th, New York became the first state in America to ban new data centers, with Governor Kathy Hochul signing a freeze on permits for hyperscale facilities for up to a full year. She cited higher power bills, water use, and grid strain as her reasons. Meanwhile, legislation is already being introduced to extend that freeze to three years. This is happening at the same time a study revealed New York has lost $11 billion in taxes due to millionaires leaving the state, and the city recently implemented rent control that has effectively killed new housing development. Compare that to Boise, Idaho, where four people started a memory chip company called Micron in the basement of a dental office back in 1978. Today, Micron employs more than 6,000 people, stands as the third largest private employer in Idaho, and just committed to a $15 billion expansion, the largest private investment in the state’s history. One town said yes 48 years ago and is still cashing that check. The next Boise could be anywhere someone decides to welcome the future, including, perhaps, the Big Island of Hawaii. The smarter move for any governor would not be a blanket freeze but a proof of concept, a small data center pilot that generates real-world data instead of relying on academic spreadsheets. Governors today have more power and agency than they may realize, and the choice between welcoming AI infrastructure or blocking it is really a choice between the future and the past.   Apple’s Price Hikes Signal the Return of On-Premise AI Apple recently raised prices across its lineup, with the Mac Studio jumping $1,300 and even entry-level MacBooks climbing $100. Tim Cook called the memory shortage a hundred-year flood, and he is not entirely wrong. DRAM and NAND prices surged roughly 60% last quarter and are projected to climb another 13 to 18% this quarter, with some analysts expecting memory costs to double again before the cycle ends. The AI hardware boom is still in its early innings, and anyone due for an upgrade should know that prices are only heading one direction. But the deeper story here is about data ownership and the return of on-premise computing. When businesses send their data into cloud-based AI platforms, those platforms can see everything. The controversy around Anthropic launching a product that competed directly with Cursor, a development tool built on top of Anthropic’s cloud, illustrated exactly why enterprises cannot afford to hand over their intellectual capital. Goldman Sachs, Merck, Citibank, none of them can afford to have an AI provider see their most sensitive work and potentially act on it. Apple’s privacy-first approach and its push to run more AI directly on device is not just a marketing position. It is a strategic response to a real problem. As LLMs commoditize, Apple is positioning itself as the gateway that routes your queries to the right model for the right task, while keeping your data on your device and out of someone else’s servers. Dell is also worth watching here, as its infrastructure business is growing at 40% while its consumer hardware grows at just 5%, a clear signal that the on-prem shift is accelerating.   The Costco Cashier Proves Category Kings Build Millionaires The Wall Street Journal ran a story about a Costco cashier who makes $32.90 an hour, started at $5.85 back when it was still Price Club, owns a three-bedroom home with a pool, and has a 401(k) worth over one million dollars. He is not an outlier. Costco’s CFO confirmed that many thousands of their hourly workers have crossed the seven-figure mark in retirement savings, and the company’s annual turnover sits at just 7% compared to a retail industry average of 60%. This story is really about category design in action. Costco became a category king in retail by capping its markups at 15% when every other retailer was charging 35 to 40%, offering generous health benefits even to part-timers, and building a culture that retains people for decades. When you combine low turnover with a growing stock, mission-driven leadership, and a business model that serves customers, employees, and investors simultaneously, you get the kind of compounding wealth that turns a cashier into a millionaire. The lesson applies whether you are scanning groceries or launching a startup. The category you pick matters more than the salary on your offer letter. Finding a company on the left side of the S-curve, one that treats its customers, its people, and its investors well while still growing, is the real career decision. The title and the paycheck matter far less than whether the category you join is heading toward abundance or quietly flatlining on the way down. To hear about all the topics in this week's The Pirate Street Journal, download and listen to this episode. You can also read more Pirate Street Journal entries in the Category Pirates newsletter.   We hope you enjoyed this episode of Christopher Lochhead: Follow Your Different™! Christopher loves hearing from his listeners. Feel free to email him, connect on Facebook, X (formerly Twitter), LinkedIn, and subscribe on Apple Podcast / Spotify!

Apple is raising its subscription prices for both Apple Music & select Apple One bundles. The OpenAI & Apple legal battle is escalating. And are Siri's AI features finally coming to China? Apple raises prices for Apple Music and Apple One subscriptions. Is Apple One worth it for you after recent price hikes? iCloud $32.8B CSAM lawsuit dismissed, Apple protected under Section 230 laws. Apple in talks to settle DOJ antitrust lawsuit, per report. Class action accuses Apple of misleading users about Hide My Email's privacy protections. OpenAI and Apple legal battle escalates, poached employees warned about data deletion. Apple Intelligence finally on the road to release in China. PrismML releases Bonsai 27B, claiming first major AI model of its size fit for iPhone. Apple's New Speech API vs Whisper: The First Real Benchmark. Apple testing 'Live Notes' AI system to record Genius Bar sessions. Apple's scrapped Mac Pro plans reportedly included a new Intel model. visionOS 27 Beta: 90-Hertz-Freeze bei 24p-Quellen und neue Controller-Plattform. A.I. 'Vibecoded' apps are flooding Apple's app store. Investigation reveals dozens of disguised gambling apps on the App Store in Brazil. San Francisco demands Apple and Google delete AI 'nudify' apps from App Stores. Picks of the Week Jason's Pick: Breville One-Touch Tea Maker Christina's Pick: Nativ Andy's Picks: David Bowie NASA Bolt Sticker & dbrand MacBook skin Hosts: Leo Laporte, Andy Ihnatko, Jason Snell, and Christina Warren Download or subscribe to MacBreak Weekly at https://twit.tv/shows/macbreak-weekly. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: ethos.com/macbreak cirasync.com/MacBreak

Apple is raising its subscription prices for both Apple Music & select Apple One bundles. The OpenAI & Apple legal battle is escalating. And are Siri's AI features finally coming to China? Apple raises prices for Apple Music and Apple One subscriptions. Is Apple One worth it for you after recent price hikes? iCloud $32.8B CSAM lawsuit dismissed, Apple protected under Section 230 laws. Apple in talks to settle DOJ antitrust lawsuit, per report. Class action accuses Apple of misleading users about Hide My Email's privacy protections. OpenAI and Apple legal battle escalates, poached employees warned about data deletion. Apple Intelligence finally on the road to release in China. PrismML releases Bonsai 27B, claiming first major AI model of its size fit for iPhone. Apple's New Speech API vs Whisper: The First Real Benchmark. Apple testing 'Live Notes' AI system to record Genius Bar sessions. Apple's scrapped Mac Pro plans reportedly included a new Intel model. visionOS 27 Beta: 90-Hertz-Freeze bei 24p-Quellen und neue Controller-Plattform. A.I. 'Vibecoded' apps are flooding Apple's app store. Investigation reveals dozens of disguised gambling apps on the App Store in Brazil. San Francisco demands Apple and Google delete AI 'nudify' apps from App Stores. Picks of the Week Jason's Pick: Breville One-Touch Tea Maker Christina's Pick: Nativ Andy's Picks: David Bowie NASA Bolt Sticker & dbrand MacBook skin Hosts: Leo Laporte, Andy Ihnatko, Jason Snell, and Christina Warren Download or subscribe to MacBreak Weekly at https://twit.tv/shows/macbreak-weekly. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: ethos.com/macbreak cirasync.com/MacBreak

Apple is raising its subscription prices for both Apple Music & select Apple One bundles. The OpenAI & Apple legal battle is escalating. And are Siri's AI features finally coming to China? Apple raises prices for Apple Music and Apple One subscriptions. Is Apple One worth it for you after recent price hikes? iCloud $32.8B CSAM lawsuit dismissed, Apple protected under Section 230 laws. Apple in talks to settle DOJ antitrust lawsuit, per report. Class action accuses Apple of misleading users about Hide My Email's privacy protections. OpenAI and Apple legal battle escalates, poached employees warned about data deletion. Apple Intelligence finally on the road to release in China. PrismML releases Bonsai 27B, claiming first major AI model of its size fit for iPhone. Apple's New Speech API vs Whisper: The First Real Benchmark. Apple testing 'Live Notes' AI system to record Genius Bar sessions. Apple's scrapped Mac Pro plans reportedly included a new Intel model. visionOS 27 Beta: 90-Hertz-Freeze bei 24p-Quellen und neue Controller-Plattform. A.I. 'Vibecoded' apps are flooding Apple's app store. Investigation reveals dozens of disguised gambling apps on the App Store in Brazil. San Francisco demands Apple and Google delete AI 'nudify' apps from App Stores. Picks of the Week Jason's Pick: Breville One-Touch Tea Maker Christina's Pick: Nativ Andy's Picks: David Bowie NASA Bolt Sticker & dbrand MacBook skin Hosts: Leo Laporte, Andy Ihnatko, Jason Snell, and Christina Warren Download or subscribe to MacBreak Weekly at https://twit.tv/shows/macbreak-weekly. Join Club TWiT for Ad-Free Podcasts! Support what you love and get ad-free audio and video feeds, a members-only Discord, and exclusive content. Join today: https://twit.tv/clubtwit Sponsors: ethos.com/macbreak cirasync.com/MacBreak

Fresh off Macstock, you get the full debrief on this year’s gathering before diving into a stacked round of Quick Tips. You’ll learn how to disable Summarize Notification Previews to stop needless battery churn, clear every unread Apple Watch notification in one shot, and weigh whether it’s finally time to move to macOS Tahoe. There’s a heads-up that Mac minis and Mac Studios are back in Apple’s Refurbished Store, a slick Control-Tilde trick to reveal formulas in Excel, Google Sheets, and LibreOffice, the case for a USB-only cable when you’re heading into DFU recovery mode, and clever moves like stretching a Kindle book loan with an old iPad, building reusable packing-list templates, and scheduling an automatic weekly reboot to keep your Macs running clean. Then you tackle your questions: setting the default audio app on your iPhone, when prepaid mobile data might be the smarter play (and when it’s not!), and a big Don’t Get Caught warning that macOS 28 is dropping support for encrypted HFS+ drives, so decrypt or reformat those external volumes before you upgrade next year. You’ll also hear how Claude Code can become your new macOS troubleshooting partner, including some honest thoughts on leaning on AI to fix your machine. Cool Stuff Found rounds it out with UDM14 for Google results minus the AI summaries, Backdrop for live macOS wallpaper, Clock Rings for tracking time in decimal, and a cheap camera lens cover that works across your MacBook, iPad, and iPhone. As always, come learn at least five new things. 00:00:00 Mac Geek Gab 1151 for Monday, July 20th, 2026 July 20th: National Fortune Cookie Day MGG Monthly Giveaway – Win a license to Mole 00:02:00 Macstock Debrief Some new-for-the-first-time attendees David Pogue Ken Case Ken Ray Paul Conaway Quick Tips 00:00:01 DLH-QT-Disable Summarize Notification Previews to save battery churn 00:06:16 Pilot Pete-QT How to clear all your unread Apple Watch notifications 00:08:34 Is it time to update to Tahoe? 00:09:25 Mac minis and Mac Studios available again in Apple's Refurb Store 00:12:31 Companies safeguarding your data 00:16:46 Marina-QT-Control-Tilde shows the formulas in Excel (and Google Sheets, and LibreOffice, but not Numbers) 00:20:03 Tony-QT-Use a USB-only cable for DFU Recovery Mode 00:21:49 Harvey-QT-Use your old iPad to extend your Kindle Book loan period 00:24:40 Todd-QT-1150-Create Packing List Templates and Sections 00:27:50 Don-QT-Schedule a weekly reboot for your Macs sudo pmset repeat restart U 05:00:00 (“U” means sUnday) 00:33:40 Pilot Pete-QT-Green light Means the Camera is On Sponsors 00:36:56 SPONSOR: Coveron. One scam can cost you everything – use code “macgeekgab” for up to 76% off at https://coveron.com/macgeekgab to safeguard your identity. 00:38:10 SPONSOR: Shopify. If you're ready to stop putting off your business and start selling, sign up for your free trial and start selling today at https://Shopify.com/MGG Your Questions Answered and Tips Shared! 00:39:33 Dave-How can I set the default audio app on my iPhone? 00:45:30 Doug-Why go prepaid for mobile data? US Mobile (allows network flexibility) Get $25 to join America’s Super Carrier, plus get 30 days free when you transfer your number! 00:53:10 Gary-DGC-macOS 28 to drop support for encrypted HFS+ drives 00:58:10 Dave-QT-Use Claude Code for macOS troubleshooting 01:03:57 Some thoughts on Troubleshooting with AI Cool Stuff Found 01:05:35 Mike-CSF-UDM14 to get Google results sans AI summaries 01:09:34 Javier-CSF-1128–Backdrop – Live MacOS Wallpaper 01:11:34 -n-Eric-CSM-Clock Rings to track time in decimal 01:15:39 Pilot Pete-CSF-Camera Lens Cover Macbook, iPad, iPhone 01:18:48 MGG 1151 Outtro MGG Monthly Giveaway Bandwidth Provided by CacheFly Pilot Pete's Aviation Podcast: So There I Was (for Aviation Enthusiasts) The Debut Film Podcast – Adam's new podcast! Dave's Business Brain (for Entrepreneurs) and Gig Gab (for Working Musicians) Podcasts MGG Merch is Available! Mac Geek Gab iOS app Mac Geek Gab YouTube Page Mac Geek Gab Live Calendar This Week's MGG Premium Contributors MGG Apple Podcasts Reviews feedback@macgeekgab.com 224-888-GEEK Active MGG Sponsors and Coupon Codes List BackBeat Media Podcast Network

The already tense relationship between Apple and OpenAI reaches a breaking point as trade secret accusations emerge, plus your hosts discuss EU regulations, the future of gaming, and public betas on the AppleInsider Podcast.Contact your hosts:@williamgallagher_ on Threads@WGallagher on TwitterWilliam's 58keys on YouTubeWilliam Gallagher on emailWes on BlueskyWes Hilliard on emailWes's blog HillitechSponsored by:CleanMyMac by MacPaw: Get Tidy Today! Try 7 days free and use  code APPLEINSIDER20 for 20% off at clnmy.com/APPLEINSIDERLinks from the Show:Apple sues OpenAI & previous VP of product design over mass IP theftApple's corporate espionage suit against OpenAI isn't the firstOpenAI shrugs, denies responsibility for trade secret theft with vague statementsOpenAI's first hardware device will be a HomePod, but don't tell them thatOpenAI blames email mixup for why it didn't respond to Apple trade theft claimsEU regulations could force an Apple Pencil upgrade in early 2027Apple Watch, Meta Glasses, AirPods get reprieve from EU replaceable battery lawM6 era will last just six months as Apple pushes for AI-focused M7OLED iPad mini may still launch before the end of 2026Madden football returns to Mac for the first time in 19 yearsBig AI acquisitions are not off the table under new CEO John TernusFirst iOS 27, macOS 27 public betas are out, but you should still be carefulSupport the show:Support the show on Patreon or Apple Podcasts to get ad-free episodes every week, access to our private Discord channel, and early release of the show! We would also appreciate a 5-star rating and review in Apple PodcastsMore AppleInsider podcastsTune in to our HomeKit Insider podcast covering the latest news, products, apps and everything HomeKit related. Subscribe in Apple Podcasts, Overcast, or just search for HomeKit Insider wherever you get your podcasts.Subscribe and listen to our AppleInsider Daily podcast for the latest Apple news Monday through Friday. You can find it on Apple Podcasts, Overcast, or anywhere you listen to podcasts.Those interested in sponsoring the show can reach out to us at: advertising@appleinsider.com (00:00) - Intro (37:08) - EU and Apple Pencil (45:45) - iPads (50:10) - M6 and M7 (56:06) - Madden Football or something ★ Support this podcast on Patreon ★