POPULARITY
Categories
398 episodes with Vit
87 episodes with Vit
153 episodes with Vit
1,093 episodes with Vit
27 episodes with Vit
76 episodes with Vit
646 episodes with Vit
40 episodes with Vit
48 episodes with Vit
29 episodes with Vit
85 episodes with Vit
12 episodes with Vit
33 episodes with Vit
40 episodes with Vit
30 episodes with Vit
31 episodes with Vit
20 episodes with Vit
22 episodes with Vit
17 episodes with Vit
19 episodes with Vit
14 episodes with Vit
14 episodes with Vit
31 episodes with Vit
13 episodes with Vit
16 episodes with Vit
18 episodes with Vit
10 episodes with Vit
19 episodes with Vit
17 episodes with Vit
8 episodes with Vit
11 episodes with Vit
167 episodes with Vit
166 episodes with Vit
9 episodes with Vit
12 episodes with Vit
15 episodes with Vit
11 episodes with Vit
13 episodes with Vit
8 episodes with Vit
9 episodes with Vit
8 episodes with Vit
8 episodes with Vit
10 episodes with Vit
15 episodes with Vit
7 episodes with Vit
18 episodes with Vit
6 episodes with Vit
11 episodes with Vit
18 episodes with Vit
12 episodes with Vit
6 episodes with Vit
17 episodes with Vit
10 episodes with Vit
6 episodes with Vit
This podcast is made possible by our listeners and viewers. If this show has brought you value, you can support it by becoming a member of The Way Forward, our platform designed to help you find the health and freedom community (people, practitioners, schools, farms, and more) near you. Your membership directly supports the podcast and the work we do.Get all the details and secure your tickets at Anarchapulco here. Use promo code thewayforward at checkout to get 10% off your ticket! Anarchism is not what they made you believe.Catherine Bleish Bonandin spent two decades in the freedom movement, got arrested twice, was profiled by the Southern Poverty Law Center, and eventually walked away from frontline activism after a police officer added her name to a monitoring list while she was raising an infant.She now produces Anarchapulco in Mexico and Liberpulco in Serbia, and partners with Liberland. She lives on Greenbriar, a consensus-based intentional community in Texas, where she homeschools her kids and teaches Game of Village.Her story starts as a Ron Paul delegate at the 2008 RNC, learning she was being handed chant cards like a movie extra. What came after is the part most podcasts skip: fusion centers, FBI provocateurs in Austin activist circles, and the slow recognition that "agitator" was a role she had to retire from.Greenbriar runs on consensus, accountability, and the four agreements. Anarchapulco went through the darkness HBO documented in The Anarchists and came out the other side. Both prove the same point Catherine kept circling back to: freedom without responsibility isn't freedom.This one runs deep on what real community looks like when nobody's making you stay.You'll Learn:[0:00] Introduction[8:02] How DHS profiled Ron Paul supporters as potentially violent militia members[24:05] The YouTube chant that put Alex Jones on notice[32:11] When your principles start getting in the way of your actual freedom[38:57] The fourth option Claire Wolfe missed: becoming a Monopoly player[46:35] Why the health freedom fight can't be won in Washington [52:44] What HBO's The Anarchist got right, and the transformation it refused to film[1:02:48] The Howard accusation, the CIA psyop email, and what provocateur behavior looks like[1:37:31] How Cherán, Mexico, kicked out the cartels and the cops, and kept them out[1:59:29] Inside Greenbrier: the consensus-based community that's lasted since 1968[2:05:53] What Liberland actually is and how Vit found land no country was claimingResources Mentioned:"Government" - The Biggest Scam in History. Exposed! | BookBlack Flags and Windmills: Hope, Anarchy, and the Common Ground Collective by scott crow | BookThe Iron Web by Larken Rose | BookFind more from Catherine:Use promo code thewayforward for 10% off Anarchapulco/Luberpulco TicketsAnarchAwakening | Get Tickets Greenbriar Community School | WebsiteFind more from Alec:Alec Zeck | Instagram | XThe Way Forward | InstagramDonate to The Way Forward here.The Way Forward is Sponsored By:PaleoValley: 100% Grass-Fed Bone Broth Protein is a nutrient-dense, easy-to-digest source of collagen and essential amino acids. Sourced from grass-fed cows, this protein powder provides the building blocks for healthy joints, skin, and gut function—without fillers or artificial ingredients. Support the show and claim 15% off your PaleoValley order!Eating well shouldn't be complicated. Dr. Cowan's Garden makes it simple to increase your daily nutrient density with their signature vegetable powders, clean pantry staples, and pasture-raised products. Family-run and committed to "beyond-organic" quality.* Offer: Use code THEWAYFORWARD for 15% off your first order.* Shop: Dr. Cowan's GardenRMDY Academy & Collective: Homeopathy Made AccessibleHigh-quality remedies and training to support natural healing. Enroll: HereExplore: Here
3 - A DJ Oti Sziget fellépésének körülményeiről tájékoztatjuk a vendégeket - vonalban Vitézy Dávid Magyarország közlekedési és beruházási minisztere by Balázsék
00:00 - 6 óra 26:16 - Kiakadtak a vásárlók: külön pénzt kérnek a hideg üdítőért egy ceglédi boltban 44:07 - A DJ Oti Sziget fellépésének körülményeiről tájékoztatjuk a vendégeket - vonalban Vitézy Dávid Magyarország közlekedési és beruházási minisztere 1:02:28 - Egy klinika felcserélte az embriókat beültetés előtt, a szülők most meghozták életük legnehezebb döntését 1:19:22 - A Zöld-foki-szigetek kapusa, Vozinha már a legendás Tom Bradyt is maga mögé utasította az Instagramon 1:33:19 - Hatalmas a fejlődés a sportorvosi módszerekben
Esperamos sua visita: Avenida Rio Doce, 217 - Ilha dos Araújos - Governador Valadares/MG - Telefone:(33) 3275-3289 Nossa Programação: Quarta-feira: 20h Culto da Vitória Sábado: 19h Culto dos Adolecentes 19h Culto Mova Jovens Domingo: 09h Escola Bíblica Dominical 19h Culto de Adoração Reuniões de Oração Segunda-feira: 19h Terça-feira: 15h Quinta-feira: 07h Domingo: 18h Visite nosso site: http://www.moriagv.com.br Visite nossas Redes Sociais: http://www.facebook.com/moriagv http://www.youtube.com/moriagv http://www.instagram.com/moriagv http://www.soundcloud.com/moriagv
Rendkívüli kormányülés lesz hétfőn, Magyar Péter meglepetést ígér a parlamentben Narratívagyár: fideszes "valóságértelmezés" egy banki felmérés alapján Eltűntek a Pride-zászlók az Erzsébet hídról, a főváros szándékos rongálásra gyanakszik Kétségbeejtő állapotokat mutat be a János Kórház pszichiátriai osztályán a 24.hu riportja Annyi embert rúgnak ki a nagy techcégek a mesterséges intelligencia miatt, hogy Kaliforniában a kormányzónak kellett közbelépni Kiderítették, melyik a legdrágább és a legolcsóbb bank Különleges autóbusz állt forgalomba Budapesten A lengyel-ukrán kapcsolatok rossz irányban haladnak? Elismert szakember lesz az Államadósság Kezelő Központ új vezetője Vitézy Dávid: Lehet vagy 45 fok a szentendrei héven, mindenkiről ömlik a víz – A miniszter keményen kiosztotta Lázár Jánost Az iráni kapus mindent fogott, második csoportmeccsén sem tudott győzni Belgium Néhány órán belül Messiék és Mbappéék is pályán – A foci-vb mai programja A hőség mellé délután még jut egy-egy zápor A további adásainkat keresd a podcast.hirstart.hu oldalunkon. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Esperamos sua visita: Avenida Rio Doce, 217 - Ilha dos Araújos - Governador Valadares/MG - Telefone:(33) 3275-3289 Nossa Programação: Quarta-feira: 20h Culto da Vitória Sábado: 19h Culto dos Adolecentes 19h Culto Mova Jovens Domingo: 09h Escola Bíblica Dominical 19h Culto de Adoração Reuniões de Oração Segunda-feira: 19h Terça-feira: 15h Quinta-feira: 07h Domingo: 18h Visite nosso site: http://www.moriagv.com.br Visite nossas Redes Sociais: http://www.facebook.com/moriagv http://www.youtube.com/moriagv http://www.instagram.com/moriagv http://www.soundcloud.com/moriagv
Esperamos sua visita: Avenida Rio Doce, 217 - Ilha dos Araújos - Governador Valadares/MG - Telefone:(33) 3275-3289 Nossa Programação: Quarta-feira: 20h Culto da Vitória Sábado: 19h Culto dos Adolecentes 19h Culto Mova Jovens Domingo: 09h Escola Bíblica Dominical 19h Culto de Adoração Reuniões de Oração Segunda-feira: 19h Terça-feira: 15h Quinta-feira: 07h Domingo: 18h Visite nosso site: http://www.moriagv.com.br Visite nossas Redes Sociais: http://www.facebook.com/moriagv http://www.youtube.com/moriagv http://www.instagram.com/moriagv http://www.soundcloud.com/moriagv
Havi 3500 forint lesz az elektromos közbicikli Budapesten, leteszteltük az e-Bubit Game over: holnap ismerteti a kormány a "fideszes mélyállam" figuráinak elmozdítását szolgáló Alaptörvény-módosítást Feszült hangulatban kezdődnek az iráni-amerikai tárgyalások Svájcban Üzenet a Kremlből: Moszkva győzni akar! Orbán Anita szerint Szijjártóék sok iratot ledaráltak Már 11 gyanúsított van a Szőlő utcai ügyben Vitézy: A Dunakeszi Járműjavító csődje miatt kellett osztrák vasúti kocsikat bérelnie a MÁV-nak Hétfőn robbanhat a politikai bomba Orbán szerint nagy bajok történtek Brüsszelben, Magyarország megadta magát Így néz ki kívülről, belülről a viszkis rabló, Ambrus Attila álomotthona: árulják az ingatlant, még medence is van a kertben Fiatalon nem láttak benne fantáziát, gyári munkásként dolgozott a focizás mellett a németek csodacseréje Kijutott az Európa-bajnokságra a magyar válogatott! Jövő héten még magasabb csúcsokat ostromol a hőség A további adásainkat keresd a podcast.hirstart.hu oldalunkon. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Todos os sábados, na programação da Rádio Taquara, é reproduzido o programa Medicina e Saúde, apresentado pelo Dr. Mauro Werb Júnior. O patrocínio é de Laboratório Bom Pastor, Proeco, Farmácia Santé, Vitória Régia e Unimed Encosta da Serra. Confira a edição deste sábado (20/06).
A ordem de serviço para implantação da quinta faixa de rolamento da Segunda Ponte, uma das principais ligações entre Cariacica, Vila Velha e Vitória, foi dada nesta semana. E as obras de ampliação terão parte dos serviços executados durante a madrugada para reduzir os impactos no trânsito. Segundo informações do governo do Estado, as intervenções serão realizadas entre 21h e 5h, de segunda-feira a sábado. Durante esse intervalo, haverá interdição total da Segunda Ponte nos dois sentidos. Como rota alternativa, os condutores deverão utilizar a Ponte Florentino Avidos. As obras devem começar em julho e a liberação total das cinco faixas para a circulação de veículos está prevista para dezembro deste ano, conforme explica o secretário de Estado de Mobilidade e Infraestrutura, Fábio Damasceno.
"Sinfonia da Cidade" é o nome do podcast realizado para a disciplina "Jornalismo Audiovisual: Lives e Webdocs", do curso latu sensu em Jornalismo Digital e Inovação da Faculdade Cásper Líbero e ministrada pelo Prof. Me. Júlio César Fernandes.A matéria é apresentada e produzida pelos estudantes Fernanda Real, Gustavo Aurélio, Paulo Henrique, Victor Bastos, Vitor Kadooka e Vitória Gomez.
Joey and Chris are back with a field recap from two days assisting the Monteith Shop's mule deer research in the Wyoming Range. Their work focused on fawn captures, a key part of understanding mule deer survival, reproduction, genetics, and how harsh or mild winters impact the herd. They explain how collared adult does, implant transmitters, satellite location "clusters," and telemetry help researchers locate birth sites and newborn fawns. Once found, each fawn is carefully processed for data like weight, measurements, and genetic samples before being safely returned. Tuesday brought a tough challenge with doe 338, who was believed to be carrying rare triplets. A premature VIT release led to a long, cold, steep hike, but with updated data and help from Ollin Company's Starlink setup, the team eventually found and processed one fawn later confirmed as hers. Wednesday went more smoothly, with Joey helping process a set of twins after a clearer VIT signal and birth site. The episode also highlights Ollin Company's live social coverage and the bigger picture of the Monteith Shop's research, which has been tracking mule deer in the region since 2013. Thank you to Ollin for the fantastic coverage.
L'IA est partout autour de nous, pour le meilleur, sans doute, mais aussi pour le pire. En la matière, son utilisation dans la guerre est peut-être celle qui pose le plus de questions. Gaza, Ukraine, Iran... l'IA facilite l'analyse de données et l'identification des cibles. Va-t-elle finir par échapper au contrôle de l'humain ? Vit-on les prémices d'une guerre sans humains ? Quid des interrogations ethniques, morales et juridiques ?
No episódio de hoje do BB Cast Agro, Rômulo Bastos Chagas, Assessor de Agronegócios do Banco do Brasil em Vitória (ES), analisa o cenário do café conilon. A entrada da nova safra brasileira pressiona os preços, enquanto as exportações seguem aquecidas e o mercado acompanha o ritmo da colheita e os desafios logísticos do setor.Destaques do episódio: ☕ Mercado em viés defensivo: o café conilon iniciou junho sob pressão nos preços, refletindo a entrada da nova safra brasileira e o aumento gradual da oferta.
Esperamos sua visita: Avenida Rio Doce, 217 - Ilha dos Araújos - Governador Valadares/MG - Telefone:(33) 3275-3289 Nossa Programação: Quarta-feira: 20h Culto da Vitória Sábado: 19h Culto dos Adolecentes 19h Culto Mova Jovens Domingo: 09h Escola Bíblica Dominical 19h Culto de Adoração Reuniões de Oração Segunda-feira: 19h Terça-feira: 15h Quinta-feira: 07h Domingo: 18h Visite nosso site: http://www.moriagv.com.br Visite nossas Redes Sociais: http://www.facebook.com/moriagv http://www.youtube.com/moriagv http://www.instagram.com/moriagv http://www.soundcloud.com/moriagv
Esperamos sua visita: Avenida Rio Doce, 217 - Ilha dos Araújos - Governador Valadares/MG - Telefone:(33) 3275-3289 Nossa Programação: Quarta-feira: 20h Culto da Vitória Sábado: 19h Culto dos Adolecentes 19h Culto Mova Jovens Domingo: 09h Escola Bíblica Dominical 19h Culto de Adoração Reuniões de Oração Segunda-feira: 19h Terça-feira: 15h Quinta-feira: 07h Domingo: 18h Visite nosso site: http://www.moriagv.com.br Visite nossas Redes Sociais: http://www.facebook.com/moriagv http://www.youtube.com/moriagv http://www.instagram.com/moriagv http://www.soundcloud.com/moriagv
Oroszországban évtizedes börtönbüntetéseket lehet kapni egyetlen Telegram-posztért vagy kis összegű utalásért is Mi lesz az olajárral Trump bejelentése után? Ezt mondják a szakértők Bejött-e a Mol-kút feeling a Corvinus Egyetem büféiben? Nagy Ervin szerint idén is lesz augusztus 20-án tűzijáték, csak rövidebb és olcsóbb Orbán Viktor a Tisza tervéről: Nem érint, rólam szól Vitézy minisztériuma megtorpedózta, 1,2 milliárdos szerződéstől esett el Tiborczék cége Orbán Anita: Automatikusan leáll Ukrajna EU-csatlakozási folyamata, ha nem teljesíti a kisebbségi jogi megállapodást „Végleg befellegzett Orbán Viktornak!” – Az Országgyűlés kétharmaddal tiltotta el a volt kormányfőt Az izraeli hadsereg korlátlan ideig marad Libanonban, Szíriában és Gázában Keddtől felfüggesztik a Bors napilap nyomtatott kiadását Elijah Just Do It: az új-zélandiak támadója két gólt lőtt Irán ellen a vb-n A focitörténet egyik legromantikusabb null-nulla: a Zöld-foki-szigetek bravúrosan kivédekezte az ötlettelen spanyolokat Már úton van, de lassan érkezik a hőség A további adásainkat keresd a podcast.hirstart.hu oldalunkon. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Palavra do Dia de Hoje - Deus Tem Seus Próprios Meios Para nos Conduzir a Vitória!
"Átnyújtotta a magántelefonszámát egy fénymásolt cetlin" – az orvosbárók terepe lett a magánszektor A Tisza-kormány elutasította az EU migrációs paktumának végrehajtását Marad a rezsicsökkentés? – kérdezte Vitályos Eszter Kapitány Istvántól, aztán elfogadta a miniszter válaszát Nemcsak hogy lesz Tusványos, de még Orbán Viktor is ott lesz Azt már lehet tudni, honnan jött Tiborczék pénze, de az még kérdés, hogy hova ment Olcsóbb lesz a piaci alapú tankolás Elfogadták az alaptörvény tizenhatodik módosítását Rendkívüli telefonhívás Trump és Putyin között: súlyos vádak hangzottak el Ukrajnáról Évente egy kisvárosnyi dolgozó tűnik el Magyarországról: csak egyetlen úton kerülhető el a katasztrófa? Navracsics Tibor elárulta, miben reménykedhet a nemzeti oldal A fehér felsőbbrendűség jelét mutathatta egy ellenőr a VAR-szobában, a Fifa vizsgálódik A vb legkisebb nemzete nem éri be a debütálással Nem is gondolnád, mire képesek a baktériumok a ködben A további adásainkat keresd a podcast.hirstart.hu oldalunkon. Hosted by Simplecast, an AdsWizz company. See pcm.adswizz.com for information about our collection and use of personal data for advertising.
Confira a reportagem de Eduardo Couto sobre o confronto entre Vitória e Porto Vitória pela Copa ES.
É Jesus que nos garante a vitória! - Pr. Harley Apolônio - Culto da Vitória [10.06.2026] by moriagv
Todos os sábados, na programação da Rádio Taquara, é reproduzido o programa Medicina e Saúde, apresentado pelo Dr. Mauro Werb Júnior. O patrocínio é de Laboratório Bom Pastor, Proeco, Farmácia Santé, Vitória Régia e Unimed Encosta da Serra. Confira a edição deste sábado (13/06).
Fredag! Det blir ett ovanligt gulligt avsnitt. Alltifrån gulliga barn till gulliga tanter. Via knark och fetischer såklart. Detta avsnitt finns också filmat på YouTube! https://youtube.com/@kafferepetpodHar du ett skvaller som fler borde få höra? Maila det till kafferepetpod@gmail.comMissa inte vår månatliga systerpodd Cigarrummet. Bli prenumerant på www.underproduktion.se/cigarrummet18:55 - Osynlig viktminskning25:00 - Att fatta emojis31:10 - Trötta pappan34:15 - Ett generöst erbjudande39:00 - The golden nugget43:15 - Mytomanvaktmästaren48:05 - Kvarglömda effekter54:10 - Vit lögn59:55 - Internationell incident Hosted on Acast. See acast.com/privacy for more information.
Kinevezése óta a 444-nek adott először nagyinterjút Vitézy Dávid közlekedési és beruházási miniszter. Stúdiónkban beszélt a Fidesz- és a Tisza-kormányok közötti különbségekről, az uniós források hazahozatalában viselt felelősségéről, Lázár János örökségéről és a Fidesz jövőjéről is.See omnystudio.com/listener for privacy information.
durée : 00:30:33 - LSD, la série documentaire - par : François da Rocha Carneiro - Comment se fabrique l'équipe de France ? Au-delà du talent individuel, sélectionneurs et joueurs doivent composer avec des critères sportifs, humains et tactiques pour bâtir un groupe soudé, capable de dépasser les ego et les tensions. - réalisation : Maryvonne Abolivier, Anahi Morales, Marie-Laure Ciboulet Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France
Vitória sobre o Egito teve um time com mudanças, mas o lateral Wesley se machucou e foi cortado. Ederson é o seu substituto. O que fica do amistoso? E o que muda com o corte de Wesley e a chegada de Ederson?SEJA MEMBRO! Seu apoio é fundamental para que o Meiocampo continue existindo e possa fazer mais. Seja membro aqui pelo Youtube! Se você ouve via podcast, clique no link na descrição para ser membro! https://www.youtube.com/channel/UCSKkF7ziXfmfjMxe9uhVyHw/joinNEWSLETTER! Nossa newsletter chega toda sexta aberta a todos com nossos textos sobre o que rolou na semana, e às terças com conteúdo apenas para assinantes: https://newsletter.meiocampo.net/Conheça o canal do Bonsa sobre Football Manager, BonsaFM: https://www.youtube.com/@BonsaFMConheça o canal do Lobo sobre games, o Próxima Fase: https://www.youtube.com/@Proxima_FaseConheça o canal de Leandro Iamin sobre a seleção brasileira, o Sarriá: https://www.youtube.com/@SarriaBrasil
00:00 - Introdução e mercados em queda livre global00:48 - Disparada dos juros americanos e economia aquecida02:08 - Criação forte de empregos nos Estados Unidos03:53 - Tarifaço dos EUA sobre o Brasil e PIX07:31 - Juros reais altos no mercado interno brasileiro10:00 - Vitória da direita nas eleições da Colômbia11:22 - Javier Milei convida IA para a Argentina14:00 - Geopolítica, tensões no Oriente Médio e petróleo15:49 - Países do Golfo buscam rotas alternativas para petróleo20:00 - IPO da SpaceX e regras do índice S&P24:13 - Anthropic, novos chips Nvidia e petróleo chinês26:06 - Google levanta bilhões para investimentos em IA29:40 - ETF da Vanguard ultrapassa um trilhão em ativos30:32 - Blackstone limita saques em fundo de crédito31:25 - SoftBank supera Toyota e investe em IA32:40 - Ouro como reserva principal de bancos centrais33:32 - Queda do Bitcoin devido ao boom da IA36:27 - Sessão de perguntas dos espectadores e encerramento
durée : 00:30:55 - Les Nuits de France Culture - par : Albane Penaranda - L'émission "Thèmes et controverses" proposait en 1961, un dialogue captivant entre Michel Foucault et Pierre Sipriot sur la folie et la raison, alors que Foucault publiait "Histoire de la folie à l'âge classique" (1ère diffusion : 20/10/1961 sur France III Nationale). - réalisation : Mathias Le Gargasson, Antoine Dhulster, Rafik Zénine, Vincent Abouchar, Emily Vallat, Hassane M'Béchour, INA Vous aimez ce podcast ? Pour écouter tous les épisodes sans limite, rendez-vous sur Radio France
Milyen furcsa sportban lennénk világelsők, milyen volt az első pár napunk a rendszerváltás után, miért lövik le egymást a rapperek gyakrabban, mint a vonósnégyes bandák, Vitézy akkor mégsem trójai faló, hogyan keressük meg a tojás két végpontját, dolgoznánk-e egy évig ingyen az ország szebb jövőjéért, pihentető nyaralás vagy metálfesztivál, megtisztulhat-e a TV2, van-e a jelnyelvnek nyelve, megbocsátunk-e a mostmár mégsem fideszes celebeknek, mire használna Magyarország egy kémműholdat, megcsalásnak számít-e egy humanoid szexrobottal való enyelgés? Valamint a feketeleves.
Magyar Péter tegnap Berlinbe ment, hogy Friedrich Merz német kancellárral találkozzon, majd ma Párizsban Emmanuel Macron francia elnökkel folytasson kétoldalú tárgyalásokat. Hegedűs Zsolt bejelentette, hogy ki lesz az új országos tiszti főorvos és az Országos Gyógyszerészeti Intézet vezetője. Az Egészségügyi Szakmai Kollégium Szülészet és Nőgyógyászati Tagozata levélben válaszolt az egészségügyi miniszter “szívhangrendelet” kapcsán feltett kérdéseire. Vitézy Dávid közlekedési és beruházási miniszter bejelentette, uniós forrásból lecserélik a HÉV-flottát. Az óbudai korrupciós üggyel kapcsolatban számos helyszínen razziáztak. Kezdje a napot a HVG hírpodcastjával!
Timestamps0:00 Intro 1:25 Vedius forgets how to do comment of the week 5:30 Vedius gets into a fight about Drafting with LEC coaches 12:17 NAVI vs. KC draft breakdowns 25:34 VIT collapse 47:26 rest of LEC discussion + Finals predictions 52:47 LCS Big G's 1:03:52 TL vs. SR 1:14:12 LCS predictions + MSI lookahead
O Santos bateu o Vitória e entrou na pausa da temporada para a Copa do Mundo fora do Z-4 do Campeonato Brasileiro, atingindo o último objetivo traçado pelo técnico Cuca. Nesta edição, Bruno Gutierrez e Nagila Luz, a Voz da Torcida, repercutem a partida, os destaques individuais do Santos e a necessidade de equilibrar o elenco para o restante da temporada. Além disso, a votação do novo estatuto social do clube e possibilidade de um novo transfer ban. Dá o play!
We're announcing AIEWF speakers this week! Take the AI Engineering Survey!Today's guest Ethan first joined us for the LS Paper Club as the lead on NVIDIA Cosmos World Model, but then joined xAI and built Grok Imagine in 3 months:He comes back on Latent Space with some nuclear hot takes: that Video Models primarily get their intelligence from LLMs, not from training on video data, and that the next frontier for truly interactive, realtime, long-horizon world models is to work on LLMs (perhaps Interaction Models as well…)Put it this way: In the near term, the next Sora won't be a better video model, but a video agent.Generative Media may more closely follow the evolution of AI coding which went from focusing on one-shot output performance and cost, to multiturn reasoning and planning models for agents and systems that can plan, edit, test, debug, and submit PRs.At a certain point, coding models got so good that the only significant next step to improve performance was handling the orchestration of these models.Now as the performance of video models increases significantly across realism, consistency, & prompt adherence while becoming more cost efficient, the next evolution of video generation may also be systems that can plan, generate, edit, critique, and iterate across an entire creative task. In this episode, Ethan joins swyx and Vibhu to unpack what it actually takes to build frontier image and video systems: data, VAEs, diffusion transformers, audio-video alignment, inference speedups, and the hidden cost of storing and moving massive video datasets. From building NVIDIA's Cosmos world model to joining xAI as Grok Imagine was being built from zero to one, Ethan He has been at the center of some of the most important work in video generation, multimodal models, and real-time world models.We go deep on Grok Imagine, how a small xAI team shipped its first multimodal video model in three months, why iteration speed matters more than almost anything in model development, and why many of the biggest gains come from fixing tiny bugs in data and training pipelines. Flipbook: The future of VideomaxxingVideo agents are almost a sure bet to be the trend in the coming year. We end with a glance at what's beyond video agents:Flipbook caused a minor sensation this year when it was released, but most treat it as a fun demo. Ethan takes it very seriously — with the speed and cost of inference coming down every year, the future of custom video JIT UI is closer than you think. We talked about why videogen models may become the front end of AI, how generative UI could replace traditional HTML/CSS, why world models need to be real-time, interactive, and long-horizon, and why the future of video generation may depend more on language models and agents than on diffusion alone.We discuss:* Why fast iteration mattered more than meetings* Why small training bugs can drive huge model quality gains* Why coding models may make compute the bottleneck again* How image and video models are trained with synthetic captions* The role of VAEs and latent space in frontier video models* Why image models are the foundation for video models* The tradeoff between temporal compression and real-time interactivity* Flipbook, Neural OS, and the future of generative UI* Why future interfaces may go from user intent to pixels* The hidden cost of training video models: storage, egress, and GPU hours* How step distillation and consistency models (like OpenAI sCM) makes video inference orders of magnitude faster* Grok Imagine 0.9 and large-scale audio-video generation* Why audio-video alignment is harder than text-video alignment* Ethan's definition of world models* Reference-to-video, video extension, and long-context video generation* Why xAI's research communication undersells Grok Imagine* How xAI culture shaped the speed of development* AI watermarking, SynthID, and detecting generated media* Why prompt rewriting matters for video models* Grok Imagine Agent and the rise of video agents* Why language models may unlock better video generation* Robotics, physical AI, and embodied world models* Why Ethan left xAI and shifted focus toward LLMs* Self-managed context, memory, and the next frontier for language modelsEthan He* LinkedIn: https://www.linkedin.com/in/ethanhe42* X: https://x.com/EthanHe_42Timestamps00:00:00 Introduction00:01:25 From NVIDIA Cosmos to xAI00:03:24 Building Grok Imagine from Zero to One00:10:07 How Image and Video Models Are Trained00:18:53 Video Compression, VAEs, and Real-Time Tradeoffs00:22:10 Generative UI, Flipbook, and Neural OS00:32:10 The Cost of Training Large Video Models00:37:04 Distillation, GANs, and Fast Video Inference00:41:21 Audio-Video Generation and Grok Imagine 0.900:48:34 What Makes a World Model?00:55:51 Reference Videos, Long Context, and Video Memory01:00:11 xAI Culture, Research, and First-Principles Building01:09:45 AI Safety, Watermarking, and Prompt Rewriting01:13:10 Video Agents and AI-Assisted Creation01:27:32 Why Language Models Unlock Better Video01:31:15 Robotics, Physical AI, and Embodied World Models01:32:38 Why Ethan Left xAI01:34:16 Self-Managed Context and the Future of LLMs01:38:43 Ethan's Career Path and Closing ThoughtsTranscriptIntroduction: Ethan He, Latent Space, and the Path to xAISwyx [00:00:00]: We're here in the studio with Ethan He, most recently of xAI. Welcome.Ethan [00:00:10]: Thank you. Glad being here.Swyx [00:00:11]: We're also here with Vibhu. you were first coming to us or joining the latent space world because you were working on Kosmos at NVIDIA, and you did a paper. We loved it. you presented it as well, so thank you for doing that.Ethan [00:00:23]: I've actually, I also presented the MoEs twice at latent space.Swyx [00:00:29]: How did you actually hear about us? Did we reach out to you? Is that how it worked?Ethan [00:00:33]: No, actually, I-- the community. Like I realized, oh, there is this online community that people talk about AI and also learn from each other through papers every week through the Paperclip. It's very nice.Ethan [00:00:49]: I learned a lot.Swyx [00:00:49]: I think three years stop. We haven't stopped even on Christmas and New Years. many weeks I want to stop but it keeps going.Vibhu [00:00:58]: No, that was good. I think you had posted that you worked on a paper, and I was “Oh, very cool. We have Paperclip. Present then.”Vibhu [00:01:04]: But I might have reached out to you after.Swyx [00:01:05]: you-- because it's an amateur club, right?Swyx [00:01:08]: so it's very unusual and but we have sometimes paper authors come by and actually explain the paper. Today we just did, the poolside paper, which was apparently very good.Vibhu [00:01:18]: Came out yesterday.Vibhu [00:01:19]: pretty interesting, right? Fully open. They talk about everything, systems. So it's a good one. We'll, we'll recommend people to read it.Swyx [00:01:25]: Bring us up to speed on your transition to xAI, ‘cause I actually don't even know when you joined. just like tell the, tell the story about the sort of transition.From NVIDIA Cosmos to xAI: Scaling Video and World ModelsEthan [00:01:34]: Before xAI, I was working on Kosmos world model as in-- at NVIDIA. So Kosmos is, it's a giant video foundation models that can-- that aims to simulate the world and for-- it serves as a foundation of-- for all of the roboticists to build on top of. There, once I built the Kosmos one, I realized as this thing also has a scaling law similar to language model, we need to scale up the video models further. that's, that's why I realized I need to move to somewhere with much more compute resources. That's how ISwyx [00:02:13]: Than NVIDIA?Vibhu [00:02:14]: The GPU rich came themselves.Vibhu [00:02:19]: And timeline-wise, when was Kosmo? It was pretty early, right? It was open world model, open paper, everything.Ethan [00:02:25]: It was end of twenty-four.Vibhu [00:02:28]: End of twenty-four.Ethan [00:02:30]: Then at mid twenty-five, I moved to xAI. At that time-- I joined about the time when xAI was about to build video models and in multi-model models. There were no infra, no data, and no model, and it just-- as a few engineers, we built it in three months and released the first model, Grok Imagine zero point nine.Ethan [00:02:55]: And since then, I keep working on video models and move more from training and to post-training of the video models. For example, like a reference to videos, kind of like the cameo feature and, video extensions. And, before I left, I worked on a world model, leading a small team to focus on the real-time long horizon video generation.Building Grok Imagine From Scratch in Three MonthsSwyx [00:03:24]: Can you give like a rough roadmap of okay, you're on a brand-new team. Grok previously was only text, or they partnered with BFL for their image gen stuff. What do you-- what are the building blocks, right? You have compute, data you can procure somewhere. Like just what are like the sequence of things that people should think about when you're setting up a new team?Vibhu [00:03:43]: actually even deeper, not just data you can procure. You guys had to go through getting the data too, right? So you shipped it pretty fast, but yeahSwyx [00:03:51]: three months is likeVibhu [00:03:52]: From everythingSwyx [00:03:52]: actually like very surprisingly fast.Ethan [00:03:55]: One thing I say like thanks to my experience at NVIDIA, ‘cause first time when we were building Kosmos together, we built it, for about a year. So this is like the second time I do it. Roughly have an idea, what to do. I say the most important thing is the talent. Everyone were very strong and clever, very close with each other towards a common goal. So that speed up things a lot. So you reduce the communication bandwidth among people, and everyone can work towards the same goal. It's, it's like every day there's not that much meetings on the calendar, like maybe like a, like a sync a day, and after that it's, it's just all building. It was pretty fun at that time.Ethan [00:04:47]: And another thing is that xAI has very strong foundations of like data inference, model inference, and the supporting there can help the model develop a lot. When I look at, training models, I don't so actually the top important thing is like how many, how many iterations can you do, per day? and the more iteration can you do, you can, you can train the model much faster. So if you have very strong infra and you have a lot of compute, you can, you can train these models in very short period of time. That can give you a much larger buffer to, for errors, and it also gives you the opportunity to spot more bugs.Iteration Speed, Compute, and Debugging Model PipelinesSwyx [00:05:46]: What is an iteration? Is it like a few hundred steps or what are youEthan [00:05:50]: Let's say just the train-training the model, like from acquire new data and maybe design new algorithms and train a new model, maybe at smaller scale orSwyx [00:06:01]: So cycle time for like any hyperparam that you're searching.Ethan [00:06:04]: Cycle time and tune to like eval this model. Is this model better than my previous iteration?Ethan [00:06:11]: SoSwyx [00:06:11]: So it's like before you, someone had already set this up that you can iterate very quickly.Ethan [00:06:15]: I think the foundation there is extremely good forDeveloping and research models.Ethan [00:06:23]: And often I find is it-- this is kind of boring, but like a lot of the improvements does not come from new algorithms. It comes from finding small bugs here and there in the data pipeline, in the, in the model training pipeline. Those give, those give the biggest boost to the model quality.Vibhu [00:06:46]: It's interesting, right? So you say it's like small team, less communication bandwidth, but also a lot of quality is like find little bugs. It seems counterintuitive, right? You have a lot of people, you can iron out more of those, but it's interesting to see the other side, right?Swyx [00:07:00]: I also wonder, have you-- do you try using LLMs to look for bugs? I don't know.Ethan [00:07:05]: I remember at that time it was mid two thousand and twenty-five, so it's the coding model wasn't quite there yet. I remem- I remember like December two thousand and twenty-five, it was extremely good. Yeah, I've been, I've been using it at that time. It's, it's helpful. sometimes it produce codes that are kind of difficult to maintain, even though like the first time it built something extremely fast. But it gave the, like a spaghetti code, thousands of lines that I couldn't maintain, and the LLM itself couldn't figure out what's, what's wrong and how to improve on top of it. But now I find it much better. Yeah, I want to bring up another point here is now coding models are much more efficient and can help us implement stuff much faster. Compute might become a bottleneck again because previously, like if you want to train a new model, say you want to generate new synthetic data and then or write a new algorithm, it might take a few weeks. And during that period of time, you don't-- you might not have experiments to run. But now you can build that thing within a few hours, then you can immediately train a model.Ethan [00:08:24]: Now you have to have enough compute to try all of the ideas. So compute might be the bottleneck of iterating speed again.Swyx [00:08:36]: yeah, I actually, honestly, I think it's like kind of a stressful job because you're “Well, I should be trying everything, and if I'm not, then I'm not doing my job well.”Vibhu [00:08:48]: there's also the stress of you're eating thousands of GPUs per hour, which is very expensive and, compute can go to other researchers.Swyx [00:08:56]: You got the daddy Elon toVibhu [00:08:57]: You got daddy Elon.Ethan [00:08:59]: It wasVibhu [00:09:00]: But there's still finite amount of compute, like you want to use it, you want to use it well, you want more of it.Ethan [00:09:06]: That was quite stressful indeed. Yeah, I think one thing is the-- with coding models now, like a lot of these jobs can be automated, which is much better. A second, it's a, it's a marathon, so you got to maintain good health and, a regular schedule.Vibhu [00:09:28]: It's, it's hard to hear that when you shift from zero to nothing in two months.Swyx [00:09:32]: and, I think obviously the culture at xAI is very famously, people work very hard. one thing I did want to dive into, in our-- in the notes that you, that you sent ahead of time, you had specific comments about the cost of Video Gen training. presumably this is on the Colossus-1, right? the two hundred megawatt cluster. Any whatever you want to just share on that.Vibhu [00:09:54]: I think there's, there's three things we're talking about, right? So there's Video Gen, there's also the Image Gen model that you put out. Do you want to like complete the, okay, so zero to one, you have a few months. Just what are the stages of create Image Gen model?Swyx [00:10:06]: Oh, yeah, maybe I got distracted.How Image and Video Models Are Trained: Synthetic Captions, Tokenizers, and VAEsVibhu [00:10:07]: Sorry. and then, from there's Video Gen, there's Audio Gen. Would love to get into those next. But what is that first few months like? So small team, a lot of bugs, iterations, but what does it look like? Do we take something off the shelf? Do we just get data compute? What's, what's the few months like? How do you go to state-art Image Gen model? How do you just start?Ethan [00:10:28]: I cannot comment specifically how xAI did, but it's, it's a quite standard process. I can draw some, examples from Cosmos. So mainly it's building a video model, you actually need to build a image model first. And building these two models, the data you need is a hundred percent synthetic pair of language and image or language to video. Because on the, on the internet, actually, the videos don't naturally associate with text. So you can say, oh, like on YouTube, you have the title and you have the description and the commentsSwyx [00:11:11]: TitleEthan [00:11:11]: of a video, but usually they're not relevant to the video itself. And say maybe like the video is a natural scene of mountains or something, and the title is, I'm so happy today.Ethan [00:11:26]: So they have they have no correlation at all. So the first step is to, you have to generate synthetic pair of language with the videos. So you gather videos from the internet, and you use a VLM to caption the videos. So that part, here's a question, like how do you, how do you gather VLM to begin with? So if there's noSwyx [00:11:55]: You, so you fuse the model, right? LikeEthan [00:11:57]: Say if there's no like VLM exists, like how do you generate the text to the beginning, right? It's, it's impossible.Swyx [00:12:04]: I see.Ethan [00:12:05]: In the beginning, it's like you ask human to describe the video as detailed as possible.For example, you ask them to describe everything, like all objects, all characters, and all interaction and dialogues in the, in the videos. So that's in the protocol of Cosmos labeling. We require the objective we give to the labelers was that you have to describe the video as detailed as possible, such that a blind person hears a blob of text can reconstruct what the video is like from their head.Swyx [00:12:43]: Video or image? You're talking about images.Ethan [00:12:44]: Video or image, either one of them.Vibhu [00:12:47]: This was pretty common when we went from clip and DALL-E, right?Vibhu [00:12:51]: It's all training on really detailed captioning of images. So same is applied to video, but insteadEthan [00:12:57]: same appliedVibhu [00:12:57]: of using multimodal model to pass in video images and write rich descriptions, you can alsoSwyx [00:13:04]: I think there's this traditional perspective of supervised, or, very highly human curated thing. I feel like there's a unlock with unsupervised, right? Where like you have enough to bootstrap that you can just throw common corpus on it or, whatever. like unsupervised vision and language pairing, right? Like where you just have, interspersed image and text and it just learns. To me, that is the VLM breakthrough that is different from the clip, different from the LM era.Ethan [00:13:36]: It's interesting to see that you kind of need both data.Ethan [00:13:41]: For example, for theSwyx [00:13:41]: You need it to bootstrap it up. YeahEthan [00:13:43]: for the generative model training, there's also usually like a small percentage of unlabeled data. So the model is instructed to generate a video without any text instruction. That can also help the model generalize. So after this stage of generative synthetic pair, so, one important common step is to train a compressor or a tokenizer of the image or videos. So because, if you train-- If you can technically, theoretically train image or video models on pure pixels, but the problem is that the, it's, it's a lot of tokens. So like one image, it's, a thousand by a thousand, it's like one million tokens, one million pixels. It's impossible to train transformer on that. So it's, you need to train a tokenizer, which can go from image to latent space and latent space back to image.Swyx [00:14:45]: That's why we named the podcast.Swyx [00:14:48]: But, basically, you're talking about vocabulary science.Ethan [00:14:50]: so vocab.Swyx [00:14:51]: And so, what is, what is imp-- like a million is impossible?Ethan [00:14:54]: In generative models, the vocab is continuous. It's a continuous space. We can think about like you map an image to a vector. It's a, it's a fixed length vector. It's sixteen or forty-eight, something like that. And then you map that vector back to the image space. And the mapping is, has-- The mapping is patch-based. So you say you haveEthan [00:15:22]: a sixteen by sixteen patch and you match, you map that patch of pixels into this latent space.Swyx [00:15:29]: We've covered thisVibhu [00:15:30]: This is like the vision transformersSwyx [00:15:32]: VAEs,Ethan [00:15:33]: VAEs.Vibhu [00:15:34]: You basically compress your input, you do your generation, you're reasoning all that generation in smaller dimension, and then you project back out.Swyx [00:15:43]: VAE is a form compression, but I think the for me, the patching thing is from VIT, right?Ethan [00:15:48]: You can make those.Swyx [00:15:49]: Literally the, yeah, the paper is titled like sixteen by sixteen is all you need. something like that. and then I think also, people make a lot of comparisons with this kind of patching with convolutions.Swyx [00:16:02]: Which is you're, you're kind of re- reconstructing the old paradigm with the new.Ethan [00:16:05]: Actually, in VAEs, there are, there are both convolution networks and transformers. You can actually do both.Ethan [00:16:14]: After this VAE, so what you've got is you've got latent space tokens and you've got the language tokens. So now the training of the diffusion transformer, usually generative models use diffusion transformers. It is actually quite standard. It's, it's very similar to how you train a language transformer models. It's not that much difference. It's just the tokens, the visual tokens in, visual tokens out. The only difference is there's a denoising process. So you train the model to unmask some of the noise. So you add, you add random noise to the visual tokens, and then you train the model to remove those noise to generate the clean tokens. Any inference, the model can iteratively remove noise from a hundred percent noise.Swyx [00:17:12]: And then there's also, to speed things along on the tech tree of diffusion, there's CFG, and then there's, there's also, latent diffusion that, there's, there's someone in there. I think, somewhere along the line, obviously, like stability and all these other guys, pioneered a lot of this, architecture. I don't know if you want to get into that or just, or do the video side up to you.Bootstrapping Video from Image Models and Temporal CompressionEthan [00:17:37]: After you train such model, such image model, the reason it's a, it's a foundation for video models is that image models are cheaper to train, and they have much denser connection between language and text. So, sorry, language and images. For example, you train a billion, you train on a billion images, and there's a mapping from the text to the image. And the cost to train the same, like the, a billion, a billion text to a billion videos, that's much more expensive because videosNaturally have more tokens than images. Because the diffusion models, their understanding of, language purely come from this mapping. So if you don't have enough mapping, so if you only train on like a ten million videos or something, there-- you might not see enough language tokens in your training, so your model does not understand human intention enough. So that's why you really-- you train-- you first train this image diffusion models, and then you bootstrap the video model from there.Swyx [00:18:53]: One thing I did want to ask, because I-- actually, I think you're, you're the first per-- video model person I've ever talked to, I think. we've, we've like talked to Luma and all those folks. There's all these tricks in video compression where basically frame by frame there's not that much difference, so actually you don't have to regenerate or save the whole frame, right? but I think MP4 compression or something else like that.Swyx [00:19:16]: is it tempting to use that? Or as far as I can tell, everyone just treats it as, “No, we would just generate every frame.” Is that roughly the state-art?Ethan [00:19:27]: There are a few different approaches. Let's say first, like you want to just directly use MP4 compression and use that as the tokens for the transformers to train, right? So people actually have tried that, but the main challenge is the latent space for the MP4 tokens were not, were not very comprehensible for the models. It's, it's extremely hard to train on that. And there's aEthan [00:20:01]: So that's why they created VAEs, which creates more continuous, latent space, so the models can understand that latent space and learn from it much easier. Even within the VAEs, there are different difficulties of the latent space. So you can imagine something the simplest, the most naive VAE is like you have an image, and you just shuffle all of the images into a, into a vector. So you don't need to train any VAEs, right? But that latent space is extremely hard for models to train on top of. That's why there are some debate on like how do you compress the tokens. So you mentioned like you can compress frame by frame. Also, you can compress, the temporal dimension.Ethan [00:20:52]: The difference is if you compress the temporal dimension, you get a much higher compression rate. Because there's temporal redundancy between frames, because, this frame and the last frame, likely they are mostly similar, so there's only some small difference. for example, I think in 12.1 VAE, they have like a eight by eight by four compression rate. So the four temporal tokens are compressed into one tokens. That can save a lot of, save a lot of the context length. If you do it frame by frame, you have to do maybe like eight by eight by one. Your context length will be four times larger. That being said, the benefit of the frame-- per frame compression, we might come back to this later, is, real-timeness and interactivity. ‘Cause if you, if you strain the output of the model, frame by frame, you can-- the model can respond to any user request immediately. So if you have like a temporal four compression, four times compression, thenSwyx [00:22:06]: It might be laggyEthan [00:22:07]: there's a lag there in nature.Swyx [00:22:10]: So you're very pilled on this. let's just go ahead and bring it up ‘cause we have the visual prepared anyway. There's some frontier applications of real-time video gen. So Flipbook is one of the examples that went viral recently, right? What is Flipbook?Real-Time Generative UI: Flipbook, Neural OS, and Diffusion Front EndsEthan [00:22:23]: Flipbook is kind of like a web brow- web browser. You can see like it has the web bro- browser UI on top. The difference is all of the UIs are generated by generative image model in real time, and anything here are fake. But you can, you can explore inside this wor- this imaginary world. Say like we-- here we have engineering the Great Pyramid. Like the model generates this for us to understand how it works, and if we want to navigate around and understand further, we can click on some of the, some of the description here, and the model will generate a new page, new subpage describing the details we want to know about.Swyx [00:23:14]: So it's basically kind of we're playing a video, but it's pausing for our next interaction, and then it just plays the next thing based on our interaction.Swyx [00:23:23]: Which is kind of cool.Vibhu [00:23:25]: and you kind of decide your story. So this was, how do you make a pyramid? levering technique seemed interesting, right? It shows how do you take Okay, I want to know what is thisSwyx [00:23:35]: The demo, the demo tweet had more animation between frames.Vibhu [00:23:38]: I think it's just skipping,Swyx [00:23:39]: Oh, it's just skipping a lot of frames.Ethan [00:23:40]: they also have a video modeVibhu [00:23:42]: It takes a lot. There's a lot of peopleEthan [00:23:42]: but, a lot of people are using it.Ethan [00:23:45]: So it's not available.Vibhu [00:23:46]: There's a live video stream. We can try,Swyx [00:23:50]: So this is an example of the kind of future that you see at the extreme. We don't-- we're obviously not in it today.Swyx [00:23:56]: But in a world where inference is completely free this is better than generating code and text?Ethan [00:24:02]: So this is, this is a final state of where Viva will be at for word model, I think. Imagine internet doesn't exist, and then you type in google.com. Like what should, what should, what should a model show you?the model can imagine something, and this is what the model imagine. And these web pages, they completely do not exist. So I think as the inference costs come down, we are going to have generative UI for everything. If you think about how the coding model works, so they write code for a web page, and they render the code might be con- converted into binary, and the binary render the pixels on the screen. So we in machine learning, every time we have some breakthrough, obviously it's, it's more intuit. So why don't we have like user instruction to the pixel directly? So the generative UI will be user intention to the pixels directly. And say like even if I want email, let's say everyone have the same interface, but I want, I want it slightly different. I want the email to show to me like a TikTok, so I can swipe left and right for the emails. And or maybe you want something else. We can have completely different things. Or like I have I'm looking at, Instagram stories, and I don't like the Like button. I always may click it. And, generative UI resolved it. So it's going to be a revolutionary replacement of the interface. So in the future, we might have much more powerfulEthan [00:25:50]: LLMs and coding models running behind the scene. And in the, in the front-end, the diffusion model will actually be the front-end to show stuff to you. That's how I imagine it.Swyx [00:26:02]: Diffusion front-end, deterministic back-end.Swyx [00:26:04]: Something like that. I find that very expensive, but,Vibhu [00:26:08]: I find it interesting you called LLMs writing code on the back end deterministic, but okay.Swyx [00:26:14]: you write it onceVibhu [00:26:15]: Compare it toSwyx [00:26:16]: And then you execute.Ethan [00:26:17]: If you think about the cost, say, let's say H100 costs $1 per hour, and if you use this eight hours a day and thirty days, so, every month you're paying this two forty, you'll actually not wanna pay for that. That's even more expensive than Cloud Code Max. But if you think about the compute costs come down like two times every year, and I think the future will likely arrive like within few years.Vibhu [00:26:49]: It's everything, right? compute cost comes down, compute gets faster, model gets smarterEthan [00:26:54]: More efficientVibhu [00:26:54]: model gets smaller.Swyx [00:26:55]: I don't know why you say two times, ‘cause I think it's like 100 times. In language models, it is roughly one hundred to a thousand times every twelve to eighteen months, for the same given level of LMSys, ELO.Vibhu [00:27:08]: That's a net of everything, right? That's model performance alongside compute. So different than just compute costs come down. But, a very interesting future.Swyx [00:27:19]: So the web designers will have to shout out that accessibility is an issue, right? how do you deal with screen readers or whatever. But yes, this is higher bandwidth storytelling than anything you can possibly generate with code, right? So I think that's the rough idea.Ethan [00:27:34]: And I'd like to add a little bit that so human naturally have the maximum bandwidth when we are looking at things, look at videos, and we also have maximum output bandwidth when we are talking. So in the future, it might be something like we talk to AI models, and the AI model responds back with a generative UI. So that would be the maximum input and output bandwidth to interact with AI models before neural link happens.Vibhu [00:28:06]: And it's also very custom, right? Some people are very visual, some people are not as visual, right? They prefer the text. But the best thing about generative UI, right, it can also be text.Swyx [00:28:17]: There's another project that we wanted to highlight, which is the Neural OS. Kinda similar idea, but here you're literally operating, simulating an operating system with a video model.Swyx [00:28:27]: and you can play Doom, you can do Firefox. I find this like mildly less impressive, obviously, because it's an OS that I can run.Swyx [00:28:37]: But here everything is imagined.Vibhu [00:28:40]: I was, used to the Command+W to close the Firefox tab. It didn't crash. That's why I saidSwyx [00:28:45]: It's too immersive.Vibhu [00:28:46]: It's, it's too immersive for me.Swyx [00:28:47]: Too immersive.Vibhu [00:28:48]: I wanted to close the tab.Vibhu [00:28:49]: But yes, I can play generated diffusion.Swyx [00:28:51]: this is shockingly fast.Swyx [00:28:54]: Because I remember there was a demo about like maybe one to two years ago. Someone tried to do the first-person shooter with a image model. There was no consistency. It was very slow. But here it looks like realistically it's-- this is Doom.Vibhu [00:29:07]: I think there's two sides to that, right? There's okay, what is running a game? The heavy part of it is actually the game engine, all the lighting, all that stuff, the graphics. This is just kind of video, right? Like we've solved consistency. This is still, it looks like a few years old image generation. There's some temporal consistency, but it's, it's kind of just images stitched together as frame video. But it's a good visual representation to pi- to picture the future you wanna see, right? that's, that's what I see in these more so.Ethan [00:29:38]: This reminds me of how the video models gets better and better. So Neural OS is kinda if you just look at it feels like it's just a crappy version of the, like the Windows we could have, right? And, but the difference is, so the model, this model is overfitted on the existing operating systems. It can generate nothing different than that. But it's actually also similar to video models. So when we are training these video model, image model, we train them on internet. There's no imaginary supernatural stuff on the internet. But once we train this model, you can prompt the model to generate something supernatural that have never existed in the data set. So if you train your Neural OS or neural computer on the standard screen recordings on the entire internet. The model can imagine completely new interface to interact with the computer.Swyx [00:30:43]: This is one of those things that is magical to me. usually generalizing out of distribution is bad, but somehow we have learned some kind of internal world model that you say, this plus, but it looks like rainbows and butterflies, it'll do it and it will kind of make sense.Swyx [00:31:03]: So yeah, that's kind of cool. Yeah, I don't know if there's any comment more on there. I do, I do wanted to, I did wanted to touch a little bit more on the model architecture stuff, which I think you were getting. It's, really fascinating. We don't get a chance to talk about this enough. So one of the papers that we covered, we've covered every annual, segment anything release. and I don't know if you follow-- you're a computer vision guy, so youEthan [00:31:26]: I knowSwyx [00:31:27]: . So they did memory attention, which is kind of interesting. And I always think, anything where you can, across the temporal dimension, keep some consistency, I think it's, very fascinating, and I don't know if Basically, does that-- the CV side bleeding into video gen side, I think is underexplored, right? we talk about it for labeling, but actually you can borrow the architecture itself.Ethan [00:31:50]: There's, there's also complete different approaches, right? you brought up the term world model, so we went from video model to world model. There is diffusion, but there's also other approaches that people are doing. So maybe we get into those after as well,?Swyx [00:32:03]: He has a whole definition of world models and stuff. I feel like we threw a lot at you. Whatever you want to comment on.Why Video Models Are Expensive: Storage, I/O, and Training ScaleEthan [00:32:10]: I think one thing that we should actually comment back on is okay, so we were talking about the steps to train image gen to video model. One thing we don't see as much of is okay, you brought up the delta in training data, right? SoEthan [00:32:24]: you won't have as much a video model might not generalize, but what is the cost of training a large video model? So we know for LLMs roughly, okay, even like the poolside thing that came out today, right? It's a Gemma level model trained on roughly forty trillion tokens at this many H200s over this much time, right? You can see what is the exact cost of that. So how many GPU hours over how much H200 costs? So how do we do the back-end math of, same thing for video models, image models. How do you, how do you kind of break that down? I can share some back-envelope calculation. So surprisingly, video models is-- the cost is very-- is comparable to language models and obviously the largest scale is language model, maybe like a medium scale to language models. I said just storing the videos alone, it costs a lot. You can, you can maybe look up on AWS or something.Ethan [00:33:20]: You really, say if you have a billion videos and let's say, let's just say like each video, like five megabyte, then you need five petabyte to just store those videos. And also remember we talk about you use a VAE to compress the videos, and you also need to store, typically you need to store those continuous feature, in-- also in your storage. That's also comparable size with the videos themselves. So just storing these videos and the features is tens of petabytes alone. And,Swyx [00:33:58]: I just, I just looked up the calculation. Five petabytes on S3 Standard is one hundred K per month.Ethan [00:34:05]: AndSwyx [00:34:05]: It's comparableEthan [00:34:05]: and you needSwyx [00:34:06]: AndEthan [00:34:06]: And then like tens of petabytes, two hundred K. And even more expensive is you have the ingress and egress.Swyx [00:34:13]: Oh, yeah.Ethan [00:34:14]: Like you-- through the internet. You have to just to download those videos, I believe it's, it's more expensive on AWS than just storing those videos.Swyx [00:34:25]: Storing, yeah.Ethan [00:34:25]: And each training runs, you probably need to pull them once. If you train multiple times, it's, it's even more than that. So it's like just storing the network, those costs is just, it would be a few, a few millions per month to just storing everything, not to mention the GPU cost.Ethan [00:34:45]: AndSwyx [00:34:45]: my side tangent, the compute rental, like GPU rental is very efficient. There's one side, okay, you can be XAI and build your data center. Should we not just build our, storage compute as well? LikeEthan [00:34:57]: Of courseSwyx [00:34:57]: cloud cost compared to just,Ethan [00:34:59]: You save so muchSwyx [00:35:00]: store. Yeah, exactly.Swyx [00:35:01]: Especially with like egress and stuff. So.Ethan [00:35:04]: That's a good idea, but it also comes to-- there are some of its own challenges.Swyx [00:35:09]: Of course, of course.Ethan [00:35:10]: like people who build the GPU data centers, they might not expect this much, storage. And yeah, people build storage, typically they just build it somewhere with just CPUs.Swyx [00:35:23]: I just looked it up. Five-- AWS only charges for egress, not ingress. Tier five for five petabytes is two hundred and thirty K.Ethan [00:35:32]: Even more expensive than the storage.Swyx [00:35:34]: But storing is per month, right? You check in, then you cannot check out. so it's so cool. It's okay. So there's that side.Ethan [00:35:41]: So the TLDR, my backhand mathSwyx [00:35:42]: Data is larger than you think. Yes.Ethan [00:35:44]: my backhand math of GPU hours times GPU cost is also very much, I'm missing some storage.Swyx [00:35:49]: You're also-- you're basically like also more IO bound than normal training.Swyx [00:35:55]: Yes. ‘Cause like data loading, so caching everything, it becomes super important.Ethan [00:36:00]: So in Cosmos, we did a lot of optimizations to make it not IO bound. So, speaking of the training, actually training the model, the GPU cost, if you look up like the open source model, how big these video models are, I think like LTX has nineteen B parameters. That's a dense model. And people are also exploring, MoEs, so it might be twenty B active and, like a hun- hundreds B, total. So that's, that's even-- that's similar size as medium-sized LLM models. And if you, if you look at number of tokens-Uh, we disclose that in Cosmos. It's also like tens of trillions of tokens on the visual tokens. So putting this together, the cost of, training these video models, it's actually comparable with LLMs. Not to mention, the infra is slightly different from LLM, so it might be less efficient to train these models.Inference Speedups: Step Distillation, Consistency Models, and GANsSwyx [00:37:04]: Do you get the benefits of traditional diffusion speed-up? So for, images, there's LCM, LoRAs for, fine-tuning. There's, there's a lot of stuff that's beenEthan [00:37:15]: Flow matching.Swyx [00:37:16]: there's flow matching. There's a lot of stuff that's been done. there's some overlap that applies to diffusion on the inference side and stuff or?Ethan [00:37:23]: so the difference-- the inference side is a completely different story.Ethan [00:37:28]: I think for the training side, it might be a little bit hard to reduce that cost. And for the inference side, the biggest gain is from the distillation of these models. You can-- It's called step distillation, slightly different from knowledge distillation in LLMs. So you-- Typically, for flow matching models, you need like 100 steps or something. Like a distortion model even need even more, like 1,000 steps to generate a good image or video. A step distillation is try to learn to generate fewer step from the model itself. It's kind of like now we-- you use the full model to generate in 100 steps, and then you take a model that only generate 10 steps and let that model to learn from the perfect one.Ethan [00:38:25]: why this workSwyx [00:38:27]: Strong to weak seemingly.Ethan [00:38:28]: It is. It's kind ofSwyx [00:38:29]: DistillationEthan [00:38:29]: kind of like strong to weak. the-- from the modeling perspective, the strong model, the teacher model is trying to model the image and videos of inter-internet, and that distribution is extremely complex. But the step distilled model is just trying to learn from the teacher. The teacher is a model, and the size is fixed, as the distribution is much simpler than the whole internet. That's the intuition I have why step distillation can work. So usually these models serve in productions, they only run in a few steps. In Cosmos, I believe we have, we have like four step and eight steps. If you do some simpler task, image-image translation, it can even run in fewer step, like one step in Cosmos Transfer.Swyx [00:39:22]: I think this is the same intuition that guides a lot of the consistency model work. I sent you a link for, SCM. I don't know if you covered that. To me, that was actually one of, the most impressive papers I've ever seen from OpenAI.Swyx [00:39:34]: That this is the unifying grand concept of consistency models. I don't know if you have any comments on this.Ethan [00:39:41]: So there are, there are a few different approaches,Swyx [00:39:46]: Oh, yeah. Here it is.Swyx [00:39:47]: Two steps versus twenty or 100 steps, whatever. It's already done.Ethan [00:39:52]: So there are, there are a few different approaches, for example, consistency model, and there are also Actually, we shouldn't forget GAN. So GAN, actually, that was, that was the OG ofSwyx [00:40:05]: OGEthan [00:40:05]: step distillation ‘cause it trained just one step to begin with. So actually, a lot of, uh-- For example, there's a distribution matching distillation which use, which uses GAN, as one of the laws for distillation. It-- GAN just tells you, “Hey, generate an image,” and thenEthan [00:40:31]: it has a discriminator to tell, is this image real or not? So the model, the model just need to learn one of the distribution, not the full distribution. Because in training, the model is asked to reconstruct the ground truth image from the internet, which is extremely hard. And in-- When you're training GAN, it's a step process. It's just a, “Hey, you generate image. Does this image look as real as the image from the internet?” Which is a much simpler task. And, yeah, combining a lot of these approaches together, people typically do that, like consistency model and distribution matching and GAN, and we can get these few step models.Audio-Video Generation and Time AlignmentSwyx [00:41:21]: Then there's one step I wanted to add, which is audio and video.Ethan [00:41:26]: So, Grok Imagine zero point nine, I believe it's, it's a first audio video transmodel deployed at a large scale. SoSwyx [00:41:39]: And that was your first model?Ethan [00:41:40]: that was, Grok Imagine's first model. It's, it's audio video, joint generation. I think the hard part is, the modality alignment, ‘cause before this transmodel, we have, we have text to video alignment. We have this, correspondence between text and video. Typically, most of the VLMs, they understand images and videos. Video's very rare, and they don't understand audio mostly. And if you look at the audio generation on the LLM side, you can talk to them perfectly fine, but if you ask them to sing a song or something, it typically is not very good. Also, they don't have, they don't have music either. The hard part is thatUh, actually audio has two component. It has like a discrete component, a continuous component. The discrete component is like the language.Ethan [00:42:44]: So when we speak, it's just, someSwyx [00:42:47]: It's an ASR issue, yeah.Ethan [00:42:49]: It's, it's text token with some characteristics, I would say.Ethan [00:42:54]: But musicSwyx [00:42:56]: I think the speech guys would disagree with this.Swyx [00:42:57]: Like disfluencies and then,Vibhu [00:43:00]: There's tones you can get angry.Ethan [00:43:01]: Well, I say largely.Ethan [00:43:03]: the mu- but the music is completely different. It's, it's very continuous, and you cannot model them like discrete tokens in language models. this is like the hard part for models is, not to mention we have to align text, video, and audio together.Ethan [00:43:26]: SoVibhu [00:43:26]: How?Ethan [00:43:28]: So significant-- some significant challenges are like-- So first, like we talk about as the VLMs, they cannot understand most of them cannot understand audio.Ethan [00:43:39]: So you have to have some way to do the synthetic data generation for audio. You have to caption the model, and that involve, that involve synthetic data and human data effort a lot. And not just surprisingly, most of the LLMs are very bad at recognizing, like the beat, tone, and the details of the of music. They can, they can give some general prediction of which song is this, but it's very hard to describe the details of the music. like we mentioned in image generation, like you have to describe image as detailed as possible so that someone blind can reconstruct that. So here is like someoneVibhu [00:44:32]: DeafEthan [00:44:32]: someone deaf can reconstruct how the music sounds like without actually listening to it. Maybe you can think of it need to have the-- or they call the script.Vibhu [00:44:49]: Subtitles, yeah.Ethan [00:44:49]: You gotta have all the details of the music, and the dialogue.Vibhu [00:44:55]: So is the challenge there typically stuff like music and audio, or is it just Like is there a baseline? Okay, there's enough data where we can understand, narration, conversation, but there's nuances in audio that's where you hit all the data issues or is it just from stage zero, you just do it all right?Ethan [00:45:15]: So one important thing is like the alignment. So the model, the model has to know like the video and audio, the, uh-- it has to have a time-based alignment, like at which time step the video and the audio token correspond to each other. But we actually don't have this kind of alignment for most of the other modalities. If you think about like text and image, text and video, they are loosely aligned. So you can, you can have a description of what's going on in the video, but you don't have to exactly, You typically don't have exact description, oh, at, time step one second like what happened?Vibhu [00:46:02]: It's veryEthan [00:46:03]: At time step two second what happenedVibhu [00:46:03]: coarse. Yeah.Swyx [00:46:05]: So what was the ideal time step? You have to oblate it, and then it's like four seconds or something.Ethan [00:46:09]: So that comes down to how you design the model to, for the model to be aware of as a time, as a time modality. So the model is like a time aware. And that's something pretty unique if you think about LLMs. So if you ask LLM to complete a task, say they, uh-- you ask them and they will say, “Oh, this task will probably take twelve hours to complete,” and they come back in one hour. Say “I've already spent two days on this and I've exhausted everything.”Ethan [00:46:47]: So the LLMs them-themselves, they don't have a sense of time there.Vibhu [00:46:53]: I actually don't think that's just them not having a sense of time. I think it's somewhat based, right?Vibhu [00:46:58]: Like you tell someone, “Okay, go work on this feature. Go implement this,” there's a general understanding you would have of how long that would take without LLMs working at LLM speed, right? So you think back like two years ago, if I tell you to like build me like a new front end for latent space, have a search bar, have all this, you'll estimate that it'll take a few days, right?Vibhu [00:47:19]: So you tell an LLM, “Go build this.” It'll take me a few days. But I think it's somewhat grounded as opposed to them not having the best-- Not saying that they have a great understanding, but I think that example is like you can see where it comes from, right? You're trained on all over the text.Swyx [00:47:35]: They're, they're trying to estimate what a human would say.Vibhu [00:47:37]: because that's what the, that's what the data kind of represents. It's not themEthan [00:47:41]: It came from the corpus on the internet. People have a estimate of how much time.Vibhu [00:47:45]: And not even just in direct like training samples, right? Just your world understanding of tokens of how long stuff takes, right? Go read a book. It'll take you a while, right?Vibhu [00:47:56]: Even if you do nothing but read a book, it takes a few days. So yeah, LLM, I read it took me a few hours.Vibhu [00:48:01]: It'll take me a few hours to go through this research. But this is a tangent.Swyx [00:48:05]: Somewhat, yeah.Swyx [00:48:06]: This is a train of thought I haven't really expressed until now is, which is basically like a full world model must also be recursive, meaning that the participant in the world model must also be aware that they have a world model. which is like this whole recursive thing down the, down the line. but yes, and that the world model can be wrong and that they need to update it and blah. Yeah. We've, argued this on the, newsletter as well, that there needs to be sort of recursive or adversarial world models.World Models: Real-Time, Long-Horizon, Interactive VideoVibhu [00:48:34]: just, to ask, how do you define world model?Swyx [00:48:38]: Oh, yeah, let's go there.Ethan [00:48:40]: SoVibhu [00:48:40]: So just for context, we talked about, video generation, and then there's a-- if you say there's a distinction between world models, what's your, what's your definition? How do you see the two?Ethan [00:48:53]: So disclaimer, I'm not going to debate, what is world model. Yeah. there are many definitions, so I'll just talk about my definition. Since I came from the multi-model, multi-model domain, so mainly talking from video. So world model is like real-time interactive long horizon videos. So there are three parts. so we-- let's talk about them one by one. So the so interaction, so we just, we just look at Facebook and neural computer. So the interaction part of it, so you, world model can allow you to interact with them through keyboard, mouse, and maybe also voice. So these all is-- all is a modality. You can, you can interact with the model, and the model should respond reasonably. Second part is real time. So once you, once, say, you move your mouse, if, say, the world model generate a game, how fast can the game respond? So if you're like professional CS: GO players- -my say, oh, you have to respond- He's beginner within sub ten milliseconds or- Yeah even less. So that's not most of the- No, sixty FPS. Let's go. Oh, three hundred FPS. Oh, five hundred FPS. Wait. okay, yeah. I didn't do the math, but yeah, okay. Uh- Yeah, three hundred FPS, that's a three millisecond. So you have to respond- Oh, s**t. Okay. YeahEthan [00:50:29]: within a millisecond. Most of the video models cannot do that. Yeah. And, but if you, say, if you have a video model that is, say, like a digital human, the response time might be more generous. Maybe typically, for real-time voice interaction, it's like two hundred millisecond. So that's, that's much more generous. But even two hundred millisecond is pretty, it is pretty tricky, ‘cause remember we mentionedEthan [00:51:01]: you have this, temporal compression coming from the VAE. So if you, if you don't compress the temporal dimension, your sequence length is going to explode. So if you want to have this real-time, real-timeness in your model, you have to do is one context problem. And the third part is long horizon, ‘cause we-- if you're not going to just play with, video games just, a few seconds, most video models only a few seconds. We're going to play with minutes, hours. The model have to be able to generate long-form content.Ethan [00:51:42]: So putting these three together, it's, real-time, long horizon interactive videos. I think the final state will be, for example, like a video, a video version of Playbook, where you can, you can interact with, a neural computer. You move your mouse, and you click on the generative interface, and it will reply to you through pixels- generating in real time. But getting there, it's, it's a very long way to get there. So one of the first step, at Grok Imagine, where I led a small world model team there, was to build video extension. So, video extension- it's the first step of interactivity. Yeah. It's, it's the first step. Yeah. So it's the first step- You have it here, video editing, yeah. Yeah. Yeah. So the first step is because, this unlocks long horizon videos. Typically, for most of the video generation models, you give it a prompt or an image as an initial frame. You generate video, that's it. That's just, one time, done. And some creators would try to, use the last frame as a first frame for the second video. It can-- sometimes it works, but if you do it a few times, it says the quality would decrease. And- It doesn't have that context- Yeah over the full video, so the temporal- Yeah, exactly. Yeah, ‘cause you only gave it the last frame, of course, right? Yeah. Exactly. And- it's actually a pretty fun hack. if you've seen like- Oh, no, he's saying something better. Yeah. And for example, like Vue, I remember Vue 3 has like a second context of the last video. It is slightly better than using the last frame, but it has the same problem-- similar problem that it, the quality would decrease. if you extend a few times to, one minute, the video quality would look much worse than the first video. Second, another problem is that the model doesn't have long-range knowledge of, what's happening before. Say, if they generate some dialogue, some, two people speaking, and their voice might change, over some time, especially if the second conditioning, it does not cover the previous context. So these are the core challenges. So the Grok Imagine video extension, it has historical context of all of the previous generated videos. It can, It has, it has the context of, who is speaking and what objects have appeared and everything, having that to generate the next video. So if we naively do this, you can imagine, just, put all of the previous history video tokens into the context. The context lens will easily explode. Especially for video models, that can be like a few, a few million context, I would imagine- context lens. Yes.Yeah.Swyx [00:54:58]: Let's run with that.Ethan [00:54:59]: for example, like in Cosmos, I think just five seconds of video is like a fifty K or sixty K number of tokens. So like if you do, if you do fifty second, that's a five hundred K tokens. If you do longer than that, easily explode. This long horizon, problem was the first step we're trying to solve world model. It turns out people, yeah, people love video extension. Like a lot, a lot of the creators love using video extension to create longer form videos. This is the part I liked that you have a, you have an intermediate step toward the final goal instead of just a straight shot to the final version very much.Swyx [00:55:48]: But I can see you have a strong vision of where we want to end up.Long Context, Redundancy, and Efficient Interactive VideoVibhu [00:55:51]: Does it seem like it's an efficiency issue? okay, we're at a few million tokens context,. If you draw the parallel to language models, we had very short context, two thousand, eight thousand, then, you scale it up one million, ten million. sure, there's effective context, but at the end of the day, it's just what's it worth? sure, there's a whole training data side. In video, it might be slightly easier ‘cause we have a hundred million token video, right? Just take a movie with the full context there. Like is this efficiency from an inference standpoint that like it's expensive, but we know how to solve it? Or like why is this not the approach? So like my broader point was on your second point of world models, you say it needs to be interactive and live, right? You should be able to play a game and see the interaction live. So one thing I see with research is a lot of what you actually serve is different than what you build, right? So we talked about distillation. You train big model, you distill it, you do quantization, speculative decoding. We do all this stuff to serve it efficiently. Should we not just have a solution, like a world model that can interact well, do inference optimization, serve it, distill it secondary, so make it real time after you solve it? So like a-- another parallel is say, continual learning, right? What we need is someone to solve it and show it works inefficiently. Give it a few years, people will make it efficient. Same thing with regular attention, right? It worked. Over a few years, people have different forms of attention, and we've scaled it to be efficient at log context,? So kind of two things there, right? One is it seems like it works. You've scaled it. Can we not just scale it a lot more efficiently over time? Do we need a separate approach if this works? And same thing with interaction, right? if we can get it done, like if we can solve some way that it works, we can solve making it more efficient from an inference standpoint later.Ethan [00:57:53]: that's actually a very good point. So in videos, there's actually a lot of redundancies. So we solve a lot of the pixel redundancy from VE, but there's more redundancy in long range and long horizon videos. Say, if a character appear in the first clip and then it disappeared, it only reappear at the end of the video, you probably don't need the-- the context, like in the middle of the generation. So you only need that character, where you need. So that's why, I helped build another feature. It's a reference video.Vibhu [00:58:36]: Is it here?Swyx [00:58:36]: is it the same model release or different one?Ethan [00:58:39]: It's a different one.Ethan [00:58:41]: You probably need to search onSwyx [00:58:43]: I'll find itEthan [00:58:43]: X reference to video.Ethan [00:58:46]: So reference video allow you to like upload up to seven images as condition and generate the video. Say, if like I want-- it can, it can be characters or objects or even scenes. Say like I want, I want condition on, Sean's selfie and holding a bladeSwyx [00:59:07]: We have a dogEthan [00:59:08]: or whatever.Swyx [00:59:08]: We put the dog in the thing.Ethan [00:59:09]: you can put them there and the video models will generate the video from and copies the context over. So that can solve a lot of the problems there, like the long context problem. It doesn't need to have a very long context, but it's-- I feel like it's an intermediate solution. The modelSwyx [00:59:29]: It's cheating.Ethan [00:59:30]: the model should be able to like selectively know, where should I draw the references. So say if I want to generate a movie, I generate it autoregressive, like a ten second at a time or something. And now this character appear, I can look back to where it first appear and, bring that back. Yeah, this one, I put the references. Yeah, that's, Optimus, Einstein myself, Annie.Vibhu [01:00:02]: Oddly enough, I used Grok Search to find it, and it pulled your LinkedIn post. But yeah we found it.Ethan [01:00:08]: Interesting.Vibhu [01:00:10]: ButxAI's Underrated Work, Culture, and WatermarkingSwyx [01:00:11]: this is a problem. This is not your fault, but like XAI doesn't communicate all this work that you do very well because they just have the model release and then that's it. But actually, these details are very good.Swyx [01:00:22]: As far as I understand, everything you just described is state-art, like no one else has done it.Vibhu [01:00:30]: A lot of-- yeah, I have a lot moreSwyx [01:00:32]: And then, and then you just put this blog post with the cookies. I'm this is not enough,?Swyx [01:00:37]: but I, obviously this is like the high level numbers that people want to know. But no, okay, soVibhu [01:00:42]: And I wonder, like part of that is also some labs don't share research into what happens. And ifSwyx [01:00:50]: No, but this is literally bragging about how good they are, right?Swyx [01:00:54]: Like, why would you not say that you are capable of extending with full context? this is not a secret sauce. This is like we did the work. yeah, I don't know.Ethan [01:01:02]: different labs have slightly different communication styles.Swyx [01:01:07]: Anyway, if anyone from XAI is listening we are always happy to help you tell your story. Yeah, okay, so you did references, and I think, I think kind of the point you're, you're making is it is sort of like a kludge, right? this is-- you can do seven, but what about 100?Swyx [01:01:23]: Right? Then you need a completely different thing.Ethan [01:01:26]: So I think it's-- this is, a mechanism to, select the context from the history, and you might not put the entire history into the context. for example, there's a paper called Frame Pack, which haveEthan [01:01:41]: a heuristic that the latest history, the last one second, I put the entire history, and the history before that, I would, compress it and makes the video smaller. So they follow this pattern, this build overall pattern that the maximum sequence length is fixed. So the further you are from the current frame, you have a smaller image. So this is just a heuristic. I think it can be more automatic. The model is aware like which history part of it can be select. So this part of the research is actually being actively, worked on by a lot of people. It's also quite interesting. I feel this is actually, this part of long context is a little bit ahead of the LLM part.Ethan [01:02:31]: So for example, like in LLMs, if you-- so contexts keep growing. Let's say if you call tool and the tool call history is extremely long, that's still in context, and keep growing, keep growing. Even if you switch the topic to something else, the whole context was there. There are some agentic harnesses that help you to, say, prune the tool results and, prune Like when you, when you query a file, only show like the top 200 lines or something. Those were very heuristic-driven.Swyx [01:03:08]: For listeners, we did a write-up on the cloud code, leak where there are eight different kinds of pruning, including like you prune the tool results and all that. So you can, you can read up on that kind of thing.Ethan [01:03:17]: I think, one breakthrough in continual learning might be like a way to automatically, manage its own context.Swyx [01:03:27]: These are all heuristics, and they will be replaced by machine learning.Ethan [01:03:30]: InterestinglyVibhu [01:03:32]: TheEthan [01:03:32]: the same thing is being researched in both LLMs and video models.Vibhu [01:03:36]: The interesting thing is also like in the paper you showed, it's actually happening at the model level, right? Compared to like language models, sure, we have base attention, but we'll do our own compression, we'll do our own pruning, which is separate from model error.Vibhu [01:03:49]: Eventually, it all just boils in, hopefully.Swyx [01:03:52]: I think this is a form of like attention, but like also know sort of reasoning attention. I feel like that's different than normal attention.Swyx [01:04:03]: Does that, does that make sense?Ethan [01:04:04]: It's, it's different in the sense that attention, not to mention, set sparse attention aside,
Análise pós-jogo da partida entre ABC x Vitória, válida pelo jogo da volta da semifinal da Copa do Nordeste 2026. Vem com a turma! Rumo ao penta ou rumo ao hexa? Independente de quantos títulos o Leão da Barra tenha na competição, o fato é que a conquista de mais um está logo à frente. […]
A pünkösdi hétvégén Kelenföldön kigyulladt villanymozdonyról volt szó, amely miatt szinte a teljes Dunántúl vonatközlekedése órákra leállt. Az okokról, és hogy mikor szűnhetnek meg a hasonló műszaki hibákból adódó nagy fennakadások a magyar vasúti közlekedésben, Andó Gergely vasúti közlekedésmérnököt kérdeztük. A műsor második részében a Ferrari új elektromos modellje kapcsán bekövetkezett 7%-os tőzsdei áresésről beszélgettünk Nagy Viktorral, a Portfolio Részvény rovatának vezető elemzőjével. Főbb részek: Intro – (00:00) Vasút – (01:49) Ferrari – (19:16) Kép forrása: Vitézy Dávid - FacebookSee omnystudio.com/listener for privacy information.
Esther Fischborn, Tomás Hammes e Luka Pumes analisam a derrota por 2 a 0 para o Vitória, que acabou com a sequência de sete jogos sem perder do Inter. Time de Pezzolano desperdiça muitas chances de gol, volta a vazar atrás e vai para a última rodada antes da parada para a Copa pressionado pela proximidade do Z-4. Aperte o play e ouça!
Análise pós-jogo da partidade entre Vitória x ABC, válida pelo jogo de ida da Semifinal da Copa do Nordeste 2026. Vem com a turma! Em jogo muito movimentado no Barradão, ABC abriu o placar e até lutou, mas depois das expulsões não teve forças pra resistir ao poderia do Vitória. Mais de 16 mil torcedores […]
Davi Brito e Kleber Bambam participaram do episódio #385 do Mundo da Luta. Em bate-papo com Ana Hissa, Marcos Luca Valentim e Vitória Lemos, os campeões do Big Brother Brasil se provocaram e falaram do duelo no FMS 8, que acontece no dia 30 de maio, em São Paulo. Dá o play!
Análise pós-jogo da partida entre Red Bull Bragantino x Vitória, válida pela 16ª rodada da Série A do Campeonato Brasileiro 2026. Vem com a turma! Muito desgastado do confronto da última quinta feira contra o Flamengo, o Vitória perde mais uma fora de casa. Time foi bem mexido para o confronto em Bragança, e com […]
Vitória sobre Mirassol mostra que há vida no Galo pós-Hulk? Quais os nomes de confiança do técnico? Quem está crescendo? Alonso está saindo e Fred chegando? Vinda do coordenador Guilherme Alves é um acerto? Com Henrique Fernandes, Izabela Baeta, Carol Leandro e Rogério Corrêa. Edição: Lavinia Aguiar. Dá o play!
Vitória por 3 a 1 sobre o Corinthians marca não só o hat-trick do centroavante como a melhor atuação do camisa 19 pelo clube. Arthur Cabral vai ser finalmente o ponto de referência ofensiva do time? Em pauta ainda a despedida de Barbosa da torcida e as razões para Danilo não ter entrado em campo. Rafa Barros, Letícia Marques, Isabelle Magalhães e Pedro Dep analisam o domingo movimentado dentro de campo e nos bastidores.
Mais um conteúdo no ar! Noite pra ficar na história do Leão da Barra! Diante de mais de 30 mil torcedores o rubro negro baiano buscou a remontada pra cima do todo poderoso Flamengo. Com direito a golaço de Erick (mais um!) e grande atuação coletiva, o Vitória chega às oitavas da Copa do Brasil. […]
Arnaldo Ribeiro, Eduardo Tironi, Mauro Cezar, José Trajano, Juca Kfouri e Danilo Lavieri debatem a eliminação do Flamengo para o Vitória na Copa do Brasil e as consequências na temporada, a troca de técnico e a nova crise do São Paulo, o Corinthians classificado, mas com Yuri Alberto admitindo sair, além da situação do Botafogo, eliminado pela Chapecoense
Jorge Natan recebe Sérgio Lobo e Thiago Lima para analisar derrota para o Vitória, com falha na pontaria aparecendo mais uma vez.
O que sobrou, 81 anos depois, da Grande Guerra Patriótica para a Rússia, do desembarque da Força Expedicionária Brasileira em Monte Castelo para o Brasil e do legado de Yalta para a ordem internacional contemporânea? Neste episódio em parceria com o Observatório Rússia e América Latina, Daniela Vieira Secches (PUC Minas/Ruslat) recebe Mariana da Gama Janot (INCT-Ineu) e Valdir da Silva Bezerra (@o_russianista), mestre em Relações Internacionais pela Universidade Estatal de São Petersburgo e organizador, com Boris Zabolotsky, do livro 80 Anos da Vitória na Grande Guerra Patriótica (Blucher, 2025). A conversa atravessa a contribuição massiva (e hoje contestada) da União Soviética para a derrota do nazifascismo, a entrada do Brasil no conflito a partir das contradições do Estado Novo e o modo como a memória da guerra foi mobilizada, na era Putin, para preencher o vácuo de identidade aberto pelo colapso soviético. No bloco de notícias, Giovana Dias Branco e Leonardo Henrique Alves de Lima Nascimento, pesquisadores do Ruslat, repercutem o mês de abril: a reaproximação Rússia-Cuba em meio à crise energética da ilha, a suspensão temporária das exportações de fertilizantes russos e seu impacto sobre o agronegócio brasileiro, o relatório sobre o treinamento de mais de mil criadores de conteúdo latino-americanos com participação da RT em espanhol, e a Holding Accountable Russian Mercenaries Act 2.0 (HARM Act 2.0), projeto bipartidário que tenta requalificar o Grupo Wagner e seus sucessores como organizações terroristas no contexto da intervenção dos EUA na Venezuela. No último bloco, Laura Schneider de Lima (PUC Minas/Ruslat) conversa com Boris Zabolotsky (Unifacs) sobre a insegurança ontológica da Rússia no pós-Guerra Fria e indica três filmes incontornáveis para pensar a guerra sem glorificá-la. Aperte o play. Quer apoiar o Chutando a Escada? Acesse chutandoaescada.com.br/apoio Mande um café usando nossa chave PIX: perguntas@chutandoaescada.com.br Comentários, críticas, sugestões? Escreva pra gente em perguntas@chutandoaescada.com.br Participaram deste episódio: Daniela Vieira Secches (PUC Minas / Ruslat), Valdir da Silva Bezerra (Ruslat), Mariana da Gama Janot (Programa de Pós-Graduação San Tiago Dantas), Giovana Dias Branco (Ruslat), Leonardo Henrique Alves de Lima Nascimento (Ruslat), Laura Schneider de Lima (Ruslat) e Boris Zabolotsky (Universidade Salvador – Unifacs / Ruslat). Capa do episódio: “Raising a flag over the Reichstag”, Yevgeny Khaldei, 2 de maio de 1945. Escute também no Spotify, no YouTube ou Apple Podcasts. Citados no episódio BEZERRA, Valdir da Silva; ZABOLOTSKY, Boris (orgs.). 80 anos da vitória na Grande Guerra Patriótica: memória, reconstrução e perspectivas. São Paulo: Blucher, 2025. Disponível em: https://www.blucher.com.br/bezerra-zabolotsky-os-80-anos-da-vitoria-na-grande-guerra-patriotica-memoria-reconstrucao-e-perspectivas. FERRAZ, Francisco César Alves. A guerra que não acabou: a reintegração social dos veteranos da Força Expedicionária Brasileira (1945-2000). 2003. Tese (Doutorado em História Social) – Universidade de São Paulo, São Paulo, 2003. Disponível em: https://repositorio.usp.br/item/001295507. VAÏSSE, Maurice. As relações internacionais desde 1945. Lisboa: Edições 70. Disponível em: https://www.estantevirtual.com.br/livro/as-relacoes-internacionais-desde-1945-HLQ-9833-000-BK. ESTADOS UNIDOS. Congresso. Câmara dos Representantes. Holding Accountable Russian Mercenaries Act 2.0 (HARM Act 2.0). Projeto de lei bipartidário, 2026. Disponível em: https://joewilson.house.gov/sites/evo-subsites/joewilson.house.gov/files/evo-media-document/wilssc_082_xml-20.pdf. KLIMOV, Elem (dir.). Vá e veja [Idi i smotri]. URSS: Mosfilm; Belarusfilm, 1985. 142 min. ROMM, Mikhail (dir.). O fascismo cotidiano [Obyknovennyy fashizm]. URSS: Mosfilm, 1965. 130 min. Documentário. BALAGOV, Kantemir (dir.). Uma mulher alta [Dylda]. Rússia: Non-Stop Production, 2019. 137 min. ASSAYAS, Olivier (dir.). O mago do Kremlin [The Wizard of the Kremlin]. França/Reino Unido, 2025. Mencionado em entrevista. Capítulos 00:00 — Abertura: 81 anos do fim da Segunda Guerra Mundial 00:04 — Valdir Bezerra: a Grande Guerra Patriótica e o legado soviético contestado 00:10 — Mariana Janot: Estado Novo, FEB e a memória disputada da participação brasileira 00:18 — Era Putin: memória, identidade nacional e renascimento militar 00:24 — O Brasil hoje: defesa, paz e o legado contra o fascismo 00:31 — Boletim Ruslat: Cuba, fertilizantes e a guerra informacional 00:37 — Leonardo Nascimento: Grupo Wagner, Venezuela e a geoeconomia do petróleo 00:44 — Boris Zabolotsky: insegurança ontológica, América Latina e três filmes contra a glorificação The post 81 anos depois: Rússia, Brasil e a memória da Segunda Guerra appeared first on Chutando a Escada.
Magnuz är docent i meteorologi. Han säger att han har radikaliserats. Nu följer han andra forskaraktivister upp på barrikaderna. Lyssna på alla avsnitt i Sveriges Radios app. Med vita labbrockar lämnar de skrivbord och laboratorium för att gå ut och bli en del av ”gatans parlament” i protest mot att de som styr inte tar klimatkrisen på allvar.En klimatforskare som tagit steget mot att bli aktivist är luftföroreningsexperten och meteorologen Magnuz Engardt. Efter mycket övervägande följde han med de andra forskaraktivisterna upp på barrikaderna som en Scientist Rebellion. Reportern Catharina Ericson Ulfves har följt honom och de andra aktivisterna under ett års tid för att förstå hur de resonerar och för att se vad de gör.Ett program från 2024.I maj 2026 dömde tingsrätten två av aktivisterna för skadegörelse. De dömdes till villkorlig dom samt ålas att betala skadestånd. Övriga 15 som greps hösten 2023 friades.Av: Catharina Ericson UlfvesProducent: Ylva LindgrenSlutmix: Jakob LalérMusiken i programmet:Trentemöller ”Miss you”Ulfves/Redman ”Withering”
Gabi Pessanha foi a convidada do episódio #384 do Mundo da Luta. Em bate-papo com Ana Hissa, Bia Figliuolo e Vitória Lemos, a faixa preta de jiu-jitsu comentou sua última conquista no campeonato brasileiro e falou do desejo de lutar no UFC BJJ.
Edgard Maciel de Sá, Cauê Rademaker e Phill analisam a atuação contra o Vitória, os motivos da irregularidade tricolor, a boa fase de JK, o futuro de Ganso e o jogo contra o Operário. DÁ O PLAY!
Estamos #NOAR! Com gol de Renato Kayzer, Vitória vence Ceará e se classifica para as semifinais da Copa do Nordeste. Vem acompanhar!
© 2020-2026 Ivy Podcast Discovery LLC
