[{"data":1,"prerenderedAt":760},["ShallowReactive",2],{"blog-post-nl-\u002Fblog\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x":3,"blog-posts-sidebar-nl":302},{"id":4,"title":5,"body":6,"categories":262,"date":288,"description":289,"extension":290,"heading":291,"image":292,"meta":293,"navigation":294,"originalUrl":295,"path":296,"seo":297,"slug":298,"stem":299,"updated":300,"__hash__":301},"blogNl\u002Fblog\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x.md","Wan2.2-videogeneratie: Paiton op AMD MI355X",{"type":7,"value":8,"toc":252},"minimark",[9,13,21,26,33,47,50,54,61,110,117,122,125,131,134,154,157,160,166,169,172,175,178,181,184,201,204,210,213,216,219,224,227,230,233,237,240,243,246,249],[10,11,12],"p",{},"Toen we Paiton voor het eerst begonnen te bouwen, was een van onze eerste aandachtsgebieden het optimaliseren van diffusiemodellen. Stable Diffusion XL was een van de eerste grote modellen waarin we lieten zien dat gefuseerde operators, efficiënte uitvoering en hardwarebewuste kernels een echt verschil konden maken.",[10,14,15,16,20],{},"Nu keren we terug naar die oorsprong. Door de groeiende belangstelling voor tekst-naar-videogeneratie hebben we ondersteuning toegevoegd voor ",[17,18,19],"strong",{},"Wan-AI\u002FWan2.2-T2V-A14B",", een groot tekst-naar-video-diffusiemodel. Dit is een belangrijke stap voor Paiton, want onze compiler- en runtimeaanpak is niet beperkt tot LLM-inferentie. Paiton optimaliseert echte AI-workloads in uiteenlopende modelfamilies, waaronder diffusie, videogeneratie en multimodale systemen.",[22,23,25],"h3",{"id":24},"benchmarkopstelling","Benchmarkopstelling",[10,27,28,29,32],{},"Voor deze vergelijking hebben we de ",[17,30,31],{},"Hugging Face Diffusers","-bibliotheek gebruikt om het Wan2.2-T2V-A14B-model uit te voeren op twee geavanceerde versnellers:",[34,35,36,42],"ul",{},[37,38,39],"li",{},[17,40,41],{},"AMD MI355X",[37,43,44],{},[17,45,46],{},"NVIDIA B200",[10,48,49],{},"Het doel was eenvoudig: de generatietijd voor hetzelfde model vergelijken met dezelfde op Diffusers gebaseerde workflow. Op de AMD MI355X gebruikten we daarvoor onze Paiton-Diffusers-plug-in.",[22,51,53],{"id":52},"resultaten","Resultaten",[10,55,56,57,60],{},"Op basis van de gemiddelde generatietijden voltooide de AMD MI355X de generatie in ",[17,58,59],{},"17,6% minder tijd"," dan de NVIDIA B200.",[62,63,64,86],"table",{},[65,66,67],"thead",{},[68,69,70,76,81],"tr",{},[71,72,73],"th",{},[17,74,75],{},"GPU",[71,77,78],{},[17,79,80],{},"Gemiddelde generatietijd",[71,82,83],{},[17,84,85],{},"Resultaat",[87,88,89,100],"tbody",{},[68,90,91,94,97],{},[92,93,46],"td",{},[92,95,96],{},"6,672s",[92,98,99],{},"Basislijn",[68,101,102,104,107],{},[92,103,41],{},[92,105,106],{},"5,501s",[92,108,109],{},"17,6% snellere generatietijd",[10,111,112],{},[113,114],"img",{"alt":115,"src":116},"","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fwan22_benhmark_evidence.jpg",[10,118,119],{},[113,120],{"alt":115,"src":121},"\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fwan22_barchart.jpg",[10,123,124],{},"Dit is precies het type workload waarvoor Paiton is gebouwd: grote modellen, zware GPU-uitvoering en dure inferentiepaden waarbij elke optimalisatie ertoe doet.",[22,126,128],{"id":127},"waarom-diffusiemodellen-er-nog-steeds-toe-doen",[17,129,130],{},"Waarom diffusiemodellen er nog steeds toe doen",[10,132,133],{},"Grote taalmodellen kregen de afgelopen jaren de meeste aandacht, maar diffusiemodellen blijven een van de belangrijkste categorieën AI-workloads. Beeldgeneratie, videogeneratie, 3D-generatie en multimodale pipelines steunen allemaal op veel van dezelfde prestatiekritische patronen:",[34,135,136,139,142,145,148,151],{},[37,137,138],{},"grote matrixbewerkingen",[37,140,141],{},"normalisatielagen",[37,143,144],{},"attentionblokken",[37,146,147],{},"geheugenzware tensortransformaties",[37,149,150],{},"herhaalde denoisingstappen",[37,152,153],{},"overhead door operatorplanning",[10,155,156],{},"Dit zijn precies de gebieden waar Paiton waarde kan bieden.",[10,158,159],{},"Ons werk aan Wan2.2-T2V-A14B bouwt voort op wat we met SDXL begonnen: complexe diffusiepipelines sneller uitvoeren met compilatie, kerneloptimalisatie en hardwarebewuste uitvoering.",[22,161,163],{"id":162},"we-streven-niet-naar-day-zero-ondersteuning",[17,164,165],{},"We streven niet naar day-zero-ondersteuning",[10,167,168],{},"We willen duidelijk zijn over onze filosofie.",[10,170,171],{},"Paiton probeert geen day-zero-ondersteuning te bieden voor elk nieuw model zodra het online verschijnt. Zo werken wij niet.",[10,173,174],{},"Wij zijn Europees. Wij nemen graag de tijd.",[10,176,177],{},"Ons doel is niet om als eerste een fragiele implementatie uit te brengen. We publiceren ondersteuning wanneer het model stabiel is, de runtime is getest en de prestatieoptimalisatie degelijk is uitgevoerd. We besteden liever meer tijd aan het begrijpen van de modelarchitectuur, het profileren van de echte knelpunten en het optimaliseren van het uitvoeringspad dan overhaast een oppervlakkige integratie uit te brengen.",[10,179,180],{},"Voor ons betekent ondersteuning niet simpelweg “het draait.”",[10,182,183],{},"Ondersteuning betekent:",[34,185,186,189,192,195,198],{},[37,187,188],{},"het model werkt correct",[37,190,191],{},"het model is goed geprofileerd",[37,193,194],{},"de dure operaties worden begrepen",[37,196,197],{},"de implementatie is geoptimaliseerd",[37,199,200],{},"het resultaat is bruikbaar in productieomgevingen",[10,202,203],{},"Dat is de standaard die we voor Paiton willen.",[22,205,207],{"id":206},"wat-dit-betekent-voor-paiton",[17,208,209],{},"Wat dit betekent voor Paiton",[10,211,212],{},"Ondersteuning voor Wan2.2-T2V-A14B is een volgende stap in de ontwikkeling van Paiton tot een breder optimalisatieplatform voor AI-inferentie.",[10,214,215],{},"We begonnen met diffusiemodellen en breidden daarna uit naar LLM's. Nu passen we die ervaring opnieuw toe op videogeneratie en moderne diffusieworkloads.",[10,217,218],{},"Het resultaat is een eenvoudige maar belangrijke boodschap:",[10,220,221],{},[17,222,223],{},"Paiton kan AMD GPU's helpen om op het hoogste niveau te concurreren op echte AI-workloads.",[10,225,226],{},"In deze benchmark behaalde de AMD MI355X, met behulp van de Diffusers-bibliotheek, een betere generatietijd dan de NVIDIA B200 op Wan2.2-T2V-A14B.",[10,228,229],{},"Voor klanten die inferentieservices bouwen voor video- of beeldgeneratie of voor grote multimodale modellen, is dit relevant. Snellere generatie betekent lagere latency, een betere benutting en lagere kosten per resultaat.",[10,231,232],{},"Paiton is precies daarvoor gebouwd.",[22,234,236],{"id":235},"bedankt-aan-amd-en-supermicro","Bedankt aan AMD en Supermicro",[10,238,239],{},"We willen AMD en Supermicro ook bedanken voor hun voortdurende ondersteuning.",[10,241,242],{},"Het bouwen van een IR-grafiekcompiler en runtimesysteem zoals Paiton vereist directe toegang tot moderne hardware, een sterk software-ecosysteem en technische ondersteuning van mensen die het platform grondig kennen. Dankzij AMD's ondersteuning konden we Paiton testen, profileren en optimaliseren op de nieuwste AMD Instinct-GPU's, waaronder de MI355X.",[10,244,245],{},"Voor een Europees bedrijf dat high-performance AI-infrastructuur bouwt, is die ondersteuning belangrijk. Ze stelt ons in staat sneller te werken, onze technologie op echte hardware te valideren en te tonen dat AMD-GPU's sterk kunnen concurreren bij veeleisende AI-workloads zoals grootschalige videogeneratie.",[10,247,248],{},"Paiton is onafhankelijke technologie, maar door toegang te hebben tot de hardware en het ecosysteem van AMD kunnen we ons optimalisatiewerk verder brengen.",[10,250,251],{},"We waarderen de samenwerking en kijken ernaar uit om door te gaan met het bouwen van hoogwaardige AI-inferentieoplossingen op AMD GPU's.",{"title":115,"searchDepth":253,"depth":253,"links":254},2,[255,257,258,259,260,261],{"id":24,"depth":256,"text":25},3,{"id":52,"depth":256,"text":53},{"id":127,"depth":256,"text":130},{"id":162,"depth":256,"text":165},{"id":206,"depth":256,"text":209},{"id":235,"depth":256,"text":236},[263,264,265,266,267,268,269,270,271,272,273,274,75,275,276,277,278,279,280,281,265,282,283,284,285,286,287],"Alle","Kunstmatige intelligentie","Paiton","14B","AMD","B200","Benchmarks","Blackwell","Compute","Diffusie","Eliovp","Generatieve AI","Hardware","Inferentie","Instinct","MI355x","NVIDIA","On-premises","Optimalisatie","Soevereine AI","T2V","Tekst-naar-video","Tuning","Video-generatie","Wan2.2","2026-06-10T14:04:04","Vergelijk Wan2.2-videogeneratie op AMD MI355X met Paiton en NVIDIA B200 via Diffusers. Lees hoe we diffusiemodellen optimaliseren.","md","Paiton keert terug naar zijn diffusiewortels: optimalisatie van Wan2.2-T2V-A14B op AMD MI355X","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaitonwan2.webp",{},true,"https:\u002F\u002Feliovp.com\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x\u002F","\u002Fblog\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x",{"title":5,"description":289},"paiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x","blog\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x",null,"cNqgJLAH9jVEZP23mjJeMogyCJZ-kUGc7swVHKxn7Bk",[303,318,328,340,351,363,372,385,387,399,421,439,458,477,495,512,524,539,554,566,575,583,598,610,621,632,642,655,665,678,689,699,710,719,731,742,751],{"path":304,"title":305,"description":306,"date":307,"slug":308,"image":309,"originalUrl":310,"categories":311},"\u002Fblog\u002Fpaiton-qwen38-mxfp4-dflash2-r9700","Qwen3.8: 400,7 tok\u002Fs op één R9700 | Paiton","Qwen3.8 op één R9700: 400,7 tok\u002Fs met ROCm 10 en vLLM 0.29, plus publieke 200K\u002F220K-chatprofielen. Benchmarks, beperkingen en startopdrachten.","2026-09-16T07:30:00Z","paiton-qwen38-mxfp4-dflash2-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen38-mxfp4\u002Fupdate-2026-09-19\u002Fhero.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-mxfp4-dflash2-r9700",[265,312,313,314,315,316,317],"AMD Radeon","Lokale AI","AI-inferentie","Inferentie-optimalisatie","Grote taalmodellen","vLLM",{"path":319,"title":320,"description":321,"date":322,"slug":323,"image":324,"originalUrl":325,"categories":326},"\u002Fblog\u002Fpaiton-qwen38-neo-gguf-vllm-r9700","Qwen3.8 GGUF in vLLM: sneller antwoord op één Radeon","Draai de originele NEO CODER MAX GGUF met Paiton in vLLM op een R9700. Bekijk de gemeten responstijden, beeldinvoer en lokale installatie.","2026-09-14T07:30:00Z","paiton-qwen38-neo-gguf-vllm-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-neo-gguf\u002F00-hero-neo-gguf-r9700.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-neo-gguf-vllm-r9700",[265,312,313,327,317],"GGUF",{"path":329,"title":330,"description":331,"date":332,"slug":333,"image":334,"originalUrl":335,"categories":336},"\u002Fblog\u002Fpaiton-minimax-h3-radeon-ai-pro-r9700","MiniMax H3 op Radeon: 15 seconden video met stereogeluid","Paiton genereert lokaal 15 seconden MiniMax H3-video met stereogeluid op één Radeon AI PRO R9700 in 5 min 33 s, met 16,7% minder wachttijd dan stock.","2026-09-09T07:30:00Z","paiton-minimax-h3-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-minimax-h3\u002F00-featured-minimax-h3-r9700.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-minimax-h3-radeon-ai-pro-r9700",[265,312,313,337,338,339],"Videogeneratie","MiniMax H3","ComfyUI",{"path":341,"title":342,"description":343,"date":344,"slug":345,"image":346,"originalUrl":347,"categories":348},"\u002Fblog\u002Fpaiton-flux2-klein-radeon-ai-pro-r9700","Lokale FLUX.2 klein op Radeon AI PRO R9700: sneller beelden genereren met minder VRAM","Paiton genereert FLUX.2 klein-beelden van 1024 × 1024 in 1,054 seconden op een R9700, met 16,2% minder generatietijd en 33,4% minder piekallocatie in Torch.","2026-09-07T09:00:00","paiton-flux2-klein-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-flux2-klein\u002Ffox-paiton.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-flux2-klein-radeon-ai-pro-r9700",[265,312,313,349,350,339],"Beeldgeneratie","FLUX",{"path":352,"title":353,"description":354,"date":355,"slug":356,"image":357,"originalUrl":358,"categories":359},"\u002Fblog\u002Fpaiton-ornith15-radeon-ai-pro-r9700","Ornith 1.5 haalt 44,6 tok\u002Fs op één Radeon AI PRO R9700","Paiton draait Ornith 1.5 35B A3B op één Radeon AI PRO R9700 met 44,63 outputtokens per seconde, 27% sneller en met 21,3% lagere gemodelleerde kosten.","2026-09-05T09:00:00","paiton-ornith15-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-ornith15\u002F00-featured-ornith15-r9700.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-ornith15-radeon-ai-pro-r9700",[265,264,312,314,360,361,315,316,317,362],"GPU-prestaties","Inferentielatentie","Kostenefficiëntie",{"path":364,"title":365,"description":366,"date":367,"slug":368,"image":369,"originalUrl":370,"categories":371},"\u002Fblog\u002Fpaiton-qwen38-radeon-ai-pro-r9700","Paiton: 21% meer Qwen3.8-doorvoer op Radeon AI PRO R9700","Paiton draait AMD's Qwen3.8 27B op één Radeon AI PRO R9700 met 39,77 outputtokens per seconde. Dat levert 21% meer throughput en 17,4% lagere gemodelleerde kosten op.","2026-09-04T09:00:00","paiton-qwen38-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-r9700\u002F00-featured-paiton-r9700.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-radeon-ai-pro-r9700",[265,264,312,314,360,361,315,316,317,362],{"path":373,"title":374,"description":375,"date":376,"slug":377,"image":378,"originalUrl":300,"categories":379},"\u002Fblog\u002Fai-data-center-power-requirements-gpu-per-megawatt","Stroomvereisten voor AI-datacenters: de GPU-per-MW-illusie","Waarom verschillen GPU-aantallen per megawatt? Lees hoe PUE, piekbelasting, opslag, netwerken en koeling de inzetbare AI-capaciteit bepalen.","2026-07-27T23:52:00","ai-data-center-power-requirements-gpu-per-megawatt","\u002Fasset\u002Fimages\u002Fblog\u002Fai-data-center-power-requirements-gpu-per-megawatt\u002Fgpu-per-megawatt-illusion.webp",[263,380,381,382,383,384],"AI-infrastructuur","Datacenters","ModFlex","HPC","AMD Helios",{"path":296,"title":5,"description":289,"date":288,"slug":298,"image":292,"originalUrl":295,"categories":386},[263,264,265,266,267,268,269,270,271,272,273,274,75,275,276,277,278,279,280,281,265,282,283,284,285,286,287],{"path":388,"title":389,"description":390,"date":391,"slug":392,"image":393,"originalUrl":394,"categories":395},"\u002Fblog\u002Ffrom-the-attic-to-the-front-page-eliovp-recognized-as-a-pioneer-in-chip-optimization-data-center-infrastructure","ElioVP in De Tijd: chipoptimalisatie en datacenters","Lees hoe De Tijd ElioVP belicht, van de oorsprong in chipoptimalisatie tot het werk aan modulaire datacenters en koeling voor hoge vermogensdichtheid.","2026-02-10T20:48:12","from-the-attic-to-the-front-page-eliovp-recognized-as-a-pioneer-in-chip-optimization-data-center-infrastructure","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fphysicalnewspaper.webp","https:\u002F\u002Feliovp.com\u002Ffrom-the-attic-to-the-front-page-eliovp-recognized-as-a-pioneer-in-chip-optimization-data-center-infrastructure\u002F",[263,264,396,397,267,398,396,279],"Modulaire DC","Niet gecategoriseerd","De Tijd",{"path":400,"title":401,"description":402,"date":403,"slug":404,"image":405,"originalUrl":406,"categories":407},"\u002Fblog\u002Fprivacy-is-geen-it-probleem-meer-het-is-een-strategische-prioriteit","AI en privacy: een strategische prioriteit in de Benelux","Privacyrisico's van generatieve AI, vertrouwen, dataopslag en governance. Waarom bedrijven in de Benelux veilige AI strategisch moeten benaderen.","2026-01-29T13:51:11","privacy-is-geen-it-probleem-meer-het-is-een-strategische-prioriteit","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fheaderimage.webp","https:\u002F\u002Feliovp.com\u002Fprivacy-is-geen-it-probleem-meer-het-is-een-strategische-prioriteit\u002F",[263,264,408,397,409,410,411,412,413,414,415,416,273,417,274,418,419,420],"Trending","AI Act","Antropomorfisme","AVG","Benelux","ChatGPT","Cyberbeveiliging","Databeheer","Gegevensbeveiliging","GDPR","Microsoft Copilot","Privacy","Shadow AI",{"path":422,"title":423,"description":424,"date":425,"slug":426,"image":427,"originalUrl":428,"categories":429},"\u002Fblog\u002Fitsme-bij-ons-is-het-its-not-me-en-dit-is-waarom","Waarom wij itsme niet gebruiken: privacy en soevereiniteit","Waarom ElioVP itsme niet gebruikt: onze afwegingen rond identiteitsmetadata, cloudafhankelijkheid, privacy en datasoevereiniteit.","2025-11-27T09:32:14","itsme-bij-ons-is-het-its-not-me-en-dit-is-waarom","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ffrontimage.webp","https:\u002F\u002Feliovp.com\u002Fitsme-bij-ons-is-het-its-not-me-en-dit-is-waarom\u002F",[263,430,431,432,414,433,434,435,417,436,437,438,419],"AWS","Belgian Mobile ID","CLOUD Act","Datasoevereiniteit","Digitale identiteit","eIDAS","itsme","Liberty Global","MyGov.be",{"path":440,"title":441,"description":442,"date":443,"slug":444,"image":445,"originalUrl":446,"categories":447},"\u002Fblog\u002Ffield-report-the-reality-of-building-agentic-ai-in-2025","Praktijkrapport: de realiteit van Agentic AI bouwen in 2025","Praktijklessen over lokale AI-agents in 2025 gaan in op workflowontwerp, observability, modeltraining, hallucinaties en GPU-geheugenlimieten.","2025-11-25T14:03:39","field-report-the-reality-of-building-agentic-ai-in-2025","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ffieldreport.webp","https:\u002F\u002Feliovp.com\u002Ffield-report-the-reality-of-building-agentic-ai-in-2025\u002F",[263,264,448,408,449,450,451,452,453,454,455,456,282,457],"Oplossingen","Agentic AI","AI-techniek","AI-strategie","Autonome agenten","Bedrijfs-AI","Lokale LLM","Modelverfijning","AI op locatie","VRAM-optimalisatie",{"path":459,"title":460,"description":461,"date":462,"slug":463,"image":464,"originalUrl":465,"categories":466},"\u002Fblog\u002Fthe-synthetic-unicorn-bubble","De synthetische unicornzeepbel","Een analyse van investeringsrisico’s bij AI-neoclouds: circulaire financiering, infrastructuurclaims, contractvoorwaarden en due diligence.","2025-11-24T19:22:28","the-synthetic-unicorn-bubble","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fsyntheticunicorn.webp","https:\u002F\u002Feliovp.com\u002Fthe-synthetic-unicorn-bubble\u002F",[263,264,408,467,468,469,470,471,472,473,474,475,476],"AI Infrastructure","AI Neocloud","Circulaire financiering","GPU Cloud","Beleggingsrisico's","Opstartwaardering","Synthetische bubbel","Technische analyse","Vaporware","Durfkapitaal",{"path":478,"title":479,"description":480,"date":481,"slug":482,"image":483,"originalUrl":484,"categories":485},"\u002Fblog\u002Fbuilding-the-engine-for-the-ai-race-the-4-month-path-to-nvidia-gb300-nvl72-power","NVIDIA GB300 NVL72: modulair datacenter in vier maanden","Ontdek een modulair datacenterontwerp voor NVIDIA GB300 NVL72, met redundante voeding, hybride koeling en een bouwplanning van vier maanden.","2025-11-20T14:10:19","building-the-engine-for-the-ai-race-the-4-month-path-to-nvidia-gb300-nvl72-power","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fsuperpodmodflexfrontimage.webp","https:\u002F\u002Feliovp.com\u002Fbuilding-the-engine-for-the-ai-race-the-4-month-path-to-nvidia-gb300-nvl72-power\u002F",[263,396,397,486,380,487,488,489,490,491,492,493,494],"150 kW-rack","DLC","Hoge dichtheid","Vloeistofkoeling","Modulair datacenter","NVIDIA Blackwell Ultra","NVIDIA GB300","NVL72","Snelle implementatie",{"path":496,"title":497,"description":498,"date":499,"slug":500,"image":501,"originalUrl":502,"categories":503},"\u002Fblog\u002Fwhy-cuda-translation-wont-unlock-amds-real-potential","CUDA-vertaling versus AMD-gerichte optimalisatie","Waarom CUDA-compatibiliteit niet hetzelfde is als AMD-prestaties: over ROCm, HIP, kerneloptimalisatie en hardwaregerichte afstemming.","2025-11-12T14:48:37","why-cuda-translation-wont-unlock-amds-real-potential","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fchatgpt-image-nov-11-2025-09_16_10-pm-1.webp","https:\u002F\u002Feliovp.com\u002Fwhy-cuda-translation-wont-unlock-amds-real-potential\u002F",[263,264,265,397,504,264,505,506,507,508,509,510,265,511],"AMD MI300X","CUDA-vertaling","FP8","GPU-optimalisatie","High-performance computing","HIP","Kerneltuning","ROCm",{"path":513,"title":514,"description":515,"date":516,"slug":517,"image":518,"originalUrl":519,"categories":520},"\u002Fblog\u002Fpaiton-the-simplest-way-to-supercharge-ai-inference","Paiton: snellere AI-inferentie in uw bestaande stack","Lees hoe Paiton aansluit op bestaande inferentiestacks, met AMD MI300X-benchmarks en vergelijkingen van prestaties per dollar.","2025-11-11T10:31:22","paiton-the-simplest-way-to-supercharge-ai-inference","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaiton-powaaah.webp","https:\u002F\u002Feliovp.com\u002Fpaiton-the-simplest-way-to-supercharge-ai-inference\u002F",[263,264,265,314,521,504,362,522,315,510,265,523,317],"AMD Instinct","Hoge throughput","SGLang",{"path":525,"title":526,"description":527,"date":528,"slug":529,"image":530,"originalUrl":531,"categories":532},"\u002Fblog\u002Fstop-overpaying-paiton-mi300x-moe-beats-h200-b200-on-1m-tokens","Paiton MoE-benchmarks: MI300X versus H200 en B200","Vergelijk Qwen3-30B-A3B MoE-benchmarks van MI300X met Paiton, H200 en B200: throughput en kosten per miljoen tokens.","2025-09-26T13:36:18","stop-overpaying-paiton-mi300x-moe-beats-h200-b200-on-1m-tokens","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fhulkvshulkpaitonwins.webp","https:\u002F\u002Feliovp.com\u002Fstop-overpaying-paiton-mi300x-moe-beats-h200-b200-on-1m-tokens\u002F",[263,264,265,533,504,534,315,535,536,46,537,265,538],"AI-benchmarks","Kosten per token","Mixture of Experts","MoE","NVIDIA H200","Qwen3",{"path":540,"title":541,"description":542,"date":543,"slug":544,"image":545,"originalUrl":546,"categories":547},"\u002Fblog\u002Fagentic-ai-but-make-it-local-from-inbox-to-insight-to-action-en","Lokale Agentic AI: van inbox naar actie","Lokale AI-agents zetten e-mails, documenten en beelden om in tickets, rapporten en acties, met modellen op maat van uw gegevens en systemen.","2025-09-16T13:09:00","agentic-ai-but-make-it-local-from-inbox-to-insight-to-action-en","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ffrontfotoblog.webp","https:\u002F\u002Feliovp.com\u002Fagentic-ai-but-make-it-local-from-inbox-to-insight-to-action-en\u002F",[263,264,448,397,449,548,549,550,551,454,456,282,552,553],"Schadedetectie","Documentverwerking","E-mailautomatisering","Factuurextractie","Ticketautomatisering","Workflowautomatisering",{"path":555,"title":556,"description":557,"date":558,"slug":559,"image":560,"originalUrl":561,"categories":562},"\u002Fblog\u002Fmi300x-fp8-data-parallel-benchmarks-8-64-gpus-h200-left-behind-b200-within-reach","MI300X FP8-benchmarks: GPU-partitionering met Paiton","Bekijk hoe Paiton presteert met Llama 3.1 8B FP8 op gepartitioneerde MI300X-GPU's, vergeleken met NVIDIA H200 en B200.","2025-07-31T13:32:57","mi300x-fp8-data-parallel-benchmarks-8-64-gpus-h200-left-behind-b200-within-reach","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fimage-2-1.webp","https:\u002F\u002Feliovp.com\u002Fmi300x-fp8-data%e2%80%91parallel-benchmarks-8-64-gpus-h200-left-behind-b200-within-reach\u002F",[263,264,265,397,563,267,268,564,565,278,279,265,317],"AI","H200","MI300X",{"path":567,"title":568,"description":569,"date":570,"slug":571,"image":572,"originalUrl":573,"categories":574},"\u002Fblog\u002Fapplicable-ai-for-businesses","Toepasbare AI voor bedrijven","Ontdek hoe ElioVP lokale AI voor bedrijfsprocessen bouwt, met modeltraining op maat en automatische schadedetectie voor de logistiek.","2025-07-09T21:35:30","applicable-ai-for-businesses","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fscherm_afbeelding-2025-07-09-om-23.30.17.webp","https:\u002F\u002Feliovp.com\u002Fapplicable-ai-for-businesses\u002F",[263,264,448,449,548,549,550,551,454,456,282,552,553],{"path":576,"title":577,"description":578,"date":579,"slug":580,"image":115,"originalUrl":581,"categories":582},"\u002Fblog\u002Fintroducing-paitons-free-evaluation-models","Maak kennis met de gratis evaluatiemodellen van Paiton","Test Paiton met gratis evaluatiemodellen voor AMD-GPU's. Vergelijk de prestaties voor tekst, beeldanalyse en beeldgeneratie met uw eigen workloads.","2025-07-07T11:26:13","introducing-paitons-free-evaluation-models","https:\u002F\u002Feliovp.com\u002Fintroducing-paitons-free-evaluation-models\u002F",[263,264,265],{"path":584,"title":585,"description":586,"date":587,"slug":588,"image":589,"originalUrl":590,"categories":591},"\u002Fblog\u002Fpaiton-dramatically-faster-startup-and-performance-for-llama-3-1-405b","Llama 3.1 405B: sneller starten met Paiton op MI300X","Bekijk Paiton-benchmarks voor Llama 3.1 405B op acht AMD MI300X-GPU's, met opstarttijd, tensorparallelisme, throughput en latency.","2025-06-12T20:15:23","paiton-dramatically-faster-startup-and-performance-for-llama-3-1-405b","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fservingscreenshot.webp","https:\u002F\u002Feliovp.com\u002Fpaiton-dramatically-faster-startup-and-performance-for-llama-3-1-405b\u002F",[263,264,265,397,314,504,592,506,593,594,595,265,596,597],"Koude start","Grafiekcompilatie","Llama 3.1 405B","LLM-optimalisatie","Opstartlatentie","Tensor-parallellisme",{"path":599,"title":600,"description":601,"date":602,"slug":603,"image":604,"originalUrl":605,"categories":606},"\u002Fblog\u002Fpaiton-fp8-beats-nvidias-h200-on-amds-mi300x","Paiton FP8 verslaat NVIDIA's H200 op AMD's MI300X","Vergelijk Paiton op AMD MI300X met NVIDIA H200 voor Llama 3.1 70B FP8: throughput, wachttijd tot het eerste token en latency bij diverse batchgroottes.","2025-06-08T19:12:40","paiton-fp8-beats-nvidias-h200-on-amds-mi300x","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fblognewfp8.webp","https:\u002F\u002Feliovp.com\u002Fpaiton-fp8-beats-nvidias-h200-on-amds-mi300x\u002F",[263,264,265,397,504,607,453,274,360,361,316,594,608,609],"Koude startoptimalisatie","Model serving","vLLM-optimalisatie",{"path":611,"title":612,"description":613,"date":614,"slug":615,"image":616,"originalUrl":617,"categories":618},"\u002Fblog\u002Fmi300x-vs-h200-vs-rx-7900-xtx-vs-tenstorrent-n300s-with-vllm","MI300X, H200, RX 7900 XTX en n300s: vLLM-benchmarks","Vergelijk MI300X, H200, RX 7900 XTX en Tenstorrent n300s met vLLM: throughput, gemodelleerde tokenkosten en hardwarebeperkingen voor Llama 3 8B.","2025-05-09T14:03:58","mi300x-vs-h200-vs-rx-7900-xtx-vs-tenstorrent-n300s-with-vllm","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fcomparisontenstor.webp","https:\u002F\u002Feliovp.com\u002Fmi300x-vs-h200-vs-rx-7900-xtx-vs-tenstorrent-n300s-with-vllm\u002F",[263,264,265,448,397,267,565,279,619,620],"RX7900XTX","tenstorrent",{"path":622,"title":623,"description":624,"date":625,"slug":626,"image":627,"originalUrl":628,"categories":629},"\u002Fblog\u002Fclusterpl-empowering-gpu-cluster-investors-with-real-world-financial-insights","ClusterP&L: financiële modellen voor GPU-clusters","Ontdek hoe ClusterP&L kosten, rendement en investeringsscenario's voor GPU-clusters modelleert, met risicoanalyses en exporteerbare rapporten.","2025-05-03T10:52:22","clusterpl-empowering-gpu-cluster-investors-with-real-world-financial-insights","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fcomparisonscenarios.webp","https:\u002F\u002Feliovp.com\u002Fclusterpl-empowering-gpu-cluster-investors-with-real-world-financial-insights\u002F",[263,264,396,448,268,564,630,279,631],"MI325X","P&L-calculator",{"path":633,"title":634,"description":635,"date":636,"slug":637,"image":638,"originalUrl":639,"categories":640},"\u002Fblog\u002Fcranking-out-faster-tokens-for-fewer-dollars-amd-mi300x-vs-nvidia-h200","AMD MI300X versus NVIDIA H200: Qwen3-32B met Paiton","Vergelijk Qwen3-32B-benchmarks op AMD MI300X met Paiton en NVIDIA H200, met resultaten voor throughput, latency en hardwarekosten.","2025-05-02T21:10:30","cranking-out-faster-tokens-for-fewer-dollars-amd-mi300x-vs-nvidia-h200","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002F3ac59a73-2466-4422-b7e5-ef2e4a8ca58e.webp","https:\u002F\u002Feliovp.com\u002Fcranking-out-faster-tokens-for-fewer-dollars-amd-mi300x-vs-nvidia-h200\u002F",[263,264,265,563,267,564,641,279,265,317],"MI300",{"path":643,"title":644,"description":645,"date":646,"slug":647,"image":648,"originalUrl":649,"categories":650},"\u002Fblog\u002Fpower-meets-precision-high-density-modular-data-center-for-nvidia-nvl-deployments-1-2-mw","Modulaire datacenters voor NVIDIA NVL: 1 tot 2 MW","Ontdek modulaire datacenterontwerpen voor NVIDIA NVL-systemen, met aandacht voor vermogen, vloeistofkoeling, redundantie en uitrolplanning.","2025-05-02T14:09:59","power-meets-precision-high-density-modular-data-center-for-nvidia-nvl-deployments-1-2-mw","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Feliovp_critical-1mw-pod_rev-2_transparent.webp","https:\u002F\u002Feliovp.com\u002Fpower-meets-precision-high-density-modular-data-center-for-nvidia-nvl-deployments-1-2-mw\u002F",[263,396,651,467,488,383,489,490,652,653,493,654],"1-2MW datacenter","NVIDIA Blackwell","NVIDIA NVL","Precisiekoeling",{"path":656,"title":657,"description":658,"date":659,"slug":660,"image":661,"originalUrl":662,"categories":663},"\u002Fblog\u002Fexamining-ai-agents-in-the-medical-field-ai-that-speaks-dicom","AI-agents in de medische wereld: AI die DICOM spreekt","Ontdek een lokale AI-agent die DICOM-gegevens opzoekt en bekijk tests van beeldmodellen met geanonimiseerde medische beelden.","2025-04-11T14:45:48","examining-ai-agents-in-the-medical-field-ai-that-speaks-dicom","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fhealthcareblog-1.webp","https:\u002F\u002Feliovp.com\u002Fexamining-ai-agents-in-the-medical-field-ai-that-speaks-dicom\u002F",[263,264,448,397,563,267,664],"Zorg",{"path":666,"title":667,"description":668,"date":669,"slug":670,"image":671,"originalUrl":672,"categories":673},"\u002Fblog\u002Feliovp-bv-your-trusted-partner-for-supply-chain-resilience-amidst-new-u-s-tariffs","Amerikaanse invoerheffingen en AI-leveringszekerheid in 2025","Lees ElioVP's visie uit april 2025 op Amerikaanse invoerheffingen en leveringszekerheid voor AI-servers, HPC-systemen en modulaire datacenters.","2025-04-04T10:01:27","eliovp-bv-your-trusted-partner-for-supply-chain-resilience-amidst-new-u-s-tariffs","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ftariffsshipping.webp","https:\u002F\u002Feliovp.com\u002Feliovp-bv-your-trusted-partner-for-supply-chain-resilience-amidst-new-u-s-tariffs\u002F",[263,408,563,267,674,675,676,677],"Invoer","Taiwan","Invoerheffingen","Trump",{"path":679,"title":680,"description":681,"date":682,"slug":683,"image":684,"originalUrl":685,"categories":686},"\u002Fblog\u002Fwhy-ai-agents-are-the-future","Waarom AI-agenten de toekomst zijn","Ontdek AI-agents voor ERP, CRM, financiën en klantondersteuning, met praktijkvoorbeelden en een traject van procesanalyse tot pilot en uitrol.","2025-03-23T22:06:59","why-ai-agents-are-the-future","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ferp2.jpeg","https:\u002F\u002Feliovp.com\u002Fwhy-ai-agents-are-the-future\u002F",[263,264,448,563,687,688],"AI-agenten","ERP",{"path":690,"title":691,"description":692,"date":693,"slug":694,"image":695,"originalUrl":696,"categories":697},"\u002Fblog\u002Fthe-rise-of-open-source-ai-model-optimization","De opkomst van open-source AI-modeloptimalisatie","Verken trends in opensource-AI-optimalisatie: kwantisatie, mixture-of-experts-modellen, hardwaregerichte afstemming, RAG en edge-AI.","2025-03-22T20:59:23","the-rise-of-open-source-ai-model-optimization","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Friseofopensource.jpeg","https:\u002F\u002Feliovp.com\u002Fthe-rise-of-open-source-ai-model-optimization\u002F",[263,264,408,698,267,75,279],"AI-nieuws",{"path":700,"title":701,"description":702,"date":703,"slug":704,"image":705,"originalUrl":706,"categories":707},"\u002Fblog\u002Fintroducing-our-benchmarking-tool-powered-by-dstack","Maak kennis met onze benchmarktool, gebouwd op dstack","Ontdek onze benchmarktool met dstack: herhaalbare vLLM-tests, automatische parameterreeksen en prestatierapporten voor lokale GPU's en de cloud.","2025-03-20T14:21:59","introducing-our-benchmarking-tool-powered-by-dstack","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fbenchmarktool.jpeg","https:\u002F\u002Feliovp.com\u002Fintroducing-our-benchmarking-tool-powered-by-dstack\u002F",[263,264,265,563,267,708,709,565,265],"benchmark","LLM",{"path":711,"title":712,"description":713,"date":714,"slug":715,"image":716,"originalUrl":717,"categories":718},"\u002Fblog\u002Foptimizing-qwq-32b-by-qwen-amd-mi300x-vs-nvidia-h200","QwQ-32B optimaliseren (door Qwen): AMD MI300X versus NVIDIA H200","Vergelijk throughput en latency van QwQ-32B op AMD MI300X met Paiton en NVIDIA H200, bij kleine batches en meer gelijktijdige aanvragen.","2025-03-19T21:41:44","optimizing-qwq-32b-by-qwen-amd-mi300x-vs-nvidia-h200","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaiton4.jpeg","https:\u002F\u002Feliovp.com\u002Foptimizing-qwq-32b-by-qwen-amd-mi300x-vs-nvidia-h200\u002F",[263,264,265],{"path":720,"title":721,"description":722,"date":723,"slug":724,"image":725,"originalUrl":726,"categories":727},"\u002Fblog\u002Feliovp-featured-on-amd-tech-talk-podcast","Eliovp te gast in de AMD Tech Talk-podcast","Beluister Elio Van Puyvelde en Jim Greene in de AMD Tech Talk-podcast over het ontstaan van ElioVP en de hardware- en softwarediensten voor AI.","2025-03-19T07:53:39","eliovp-featured-on-amd-tech-talk-podcast","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Ftechtalkjimgreene.jpeg","https:\u002F\u002Feliovp.com\u002Feliovp-featured-on-amd-tech-talk-podcast\u002F",[263,267,728,729,730],"Jim Greene","Podcast","Tech Talk",{"path":732,"title":733,"description":734,"date":735,"slug":736,"image":737,"originalUrl":738,"categories":739},"\u002Fblog\u002Ffurther-optimizing-amd-powered-inference-with-paiton","AMD-inferentie verder optimaliseren met Paiton","Bekijk Paiton-benchmarks voor DeepSeek R1 Distill Llama 8B op AMD MI300X, gericht op throughput en latency bij kleinere batchgroottes.","2025-03-13T06:18:30","further-optimizing-amd-powered-inference-with-paiton","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaitonpost3.webp","https:\u002F\u002Feliovp.com\u002Ffurther-optimizing-amd-powered-inference-with-paiton\u002F",[263,264,265,267,740,741,564,565,630,265,317],"DeepSeek","H100",{"path":743,"title":744,"description":745,"date":746,"slug":747,"image":748,"originalUrl":749,"categories":750},"\u002Fblog\u002Fa-first-look-at-paiton-in-action-deepseek-r1-distill-llama-3-1-8b","Paiton-benchmarks: DeepSeek R1 Distill Llama 3.1 8B","Vergelijk standaard- en Paiton-versies van DeepSeek R1 Distill Llama 3.1 8B op AMD MI300X, met benchmarks voor throughput en latency.","2025-01-31T09:11:02","a-first-look-at-paiton-in-action-deepseek-r1-distill-llama-3-1-8b","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaitonpost2.webp","https:\u002F\u002Feliovp.com\u002Fa-first-look-at-paiton-in-action-deepseek-r1-distill-llama-3-1-8b\u002F",[263,264,265,267,740,741,564,565,630,265,317],{"path":752,"title":753,"description":754,"date":755,"slug":756,"image":757,"originalUrl":758,"categories":759},"\u002Fblog\u002Fai-model-optimization-with-paiton","AI-modeloptimalisatie met Paiton","Lees hoe Paiton modelcompilatie, aangepaste kernels en kernelfusie inzet om AI-inferentie op AMD GPU's te optimaliseren.","2025-01-30T19:53:25","ai-model-optimization-with-paiton","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaitonpost1.webp","https:\u002F\u002Feliovp.com\u002Fai-model-optimization-with-paiton\u002F",[263,264,265,267,741,564,565,630,265,317],1789853167837]