[{"data":1,"prerenderedAt":1464},["ShallowReactive",2],{"blog-post-nl-\u002Fblog\u002Fpaiton-idol-radeon-ai-pro-r9700":3,"blog-posts-sidebar-nl":898},{"id":4,"title":5,"body":6,"categories":880,"date":884,"description":885,"extension":886,"heading":887,"image":888,"meta":889,"navigation":551,"originalUrl":890,"path":891,"seo":892,"slug":893,"socialImage":894,"stem":895,"updated":896,"video":896,"__hash__":897},"blogNl\u002Fblog\u002Fpaiton-idol-radeon-ai-pro-r9700.md","De Radeon AI PRO R9700 doet ook 3D: Idol in 16,4 seconden",{"type":7,"value":8,"toc":871},"minimark",[9,19,46,49,54,82,87,94,107,119,122,126,132,137,201,214,232,255,265,269,275,280,369,384,389,395,400,480,493,496,500,522,693,707,725,733,742,753,771,867],[10,11,12,13,18],"p",{},"We hebben al veel tijd gestoken in ",[14,15,17],"a",{"href":16},"\u002Fnl\u002Fblog\u002Fpaiton-qwen38-mxfp4-dflash2-r9700","lokale taalmodellen op de R9700",". Maar een werkstation hoeft niet bij chat en code te stoppen. Wat als dezelfde kaart een afbeelding kan omzetten in een object dat u van alle kanten kunt bekijken?",[10,20,21,22,26,27,30,31,34,35],{},"Dat doet ",[23,24,25],"strong",{},"Idol van Sperid Labs",". Geef het één afbeelding en het genereert een 3D Gaussian splat, inclusief een aannemelijke achterkant die nooit op het beeld stond. Met Paiton duurt een volledig verzoek ",[23,28,29],{},"16,4 seconden op één Radeon AI PRO R9700",". Standaard-Idol heeft op dezelfde kaart 39,2 seconden nodig: ongeveer ",[23,32,33],{},"2,4× sneller",", met dezelfde gewichten, 50 stappen en guidance-instelling.",[36,37,38],"sup",{},[14,39,45],{"href":40,"ariaDescribedBy":41,"dataFootnoteRef":43,"id":44},"#user-content-fn-bench",[42],"footnote-label","","user-content-fnref-bench","1",[10,47,48],{},"Ter vergelijking: Sperid Labs rapporteert afzonderlijk 19,1 seconden op een RTX 4090. Een interessant referentiepunt, geen gecontroleerde vergelijking tussen GPU-fabrikanten. Hieronder laten we zien wat we maten, wat eruit komt en hoe u het zelf probeert.",[10,50,51],{},[23,52,53],{},"In dit artikel",[55,56,57,64,70,76],"ul",{},[58,59,60],"li",{},[14,61,63],{"href":62},"#van-beeld-naar-object","Van één afbeelding naar een draaibaar object",[58,65,66],{},[14,67,69],{"href":68},"#van-392-naar-164-seconden","De meting op dezelfde GPU en de kwaliteitscontrole",[58,71,72],{},[14,73,75],{"href":74},"#waar-de-r9700-uitkomt","Andere GPUs en een indicatieve prijsvergelijking",[58,77,78],{},[14,79,81],{"href":80},"#probeer-het-op-uw-r9700","Zelf lokaal draaien",[83,84,86],"h2",{"id":85},"van-beeld-naar-object","Van beeld naar object",[10,88,89],{},[90,91],"img",{"alt":92,"src":93},"Een boerenfiguur: links de invoerafbeelding, rechts een roterend aanzicht van de gegenereerde 3D Gaussian splat.","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fturntable-farmer.gif",[10,95,96],{},[97,98,99,100,106],"em",{},"De header is een illustratie; deze animatie toont een echte uitvoer van de uitgebrachte Paiton-bundel op één R9700, met 262.144 Gaussians. Links: invoer gegenereerd met FLUX.2 klein 4B. Rechts: het resultaat van Idol van alle kanten. ",[14,101,105],{"href":102,"rel":103},"https:\u002F\u002Fgithub.com\u002FEliovp-BV\u002Fpaiton-vllm-plugin\u002Fblob\u002F9ff80088fb0063fe0dc140200c21ebc262594aac\u002Fmodels\u002FIdol\u002FREADME.md",[104],"nofollow","Origineel voorbeeld en herkomst",".",[10,108,109,110,114,115,118],{},"De uitvoer is een standaard 3DGS-",[111,112,113],"code",{},".ply",", ",[23,116,117],{},"geen getextureerde polygonenmesh",". Open het bestand in een Gaussian-splatviewer om het object te verkennen. De onzichtbare kanten zijn gegenereerd, niet teruggewonnen metingen van het oorspronkelijke object. Dat maakt dit interessant voor visuele ideeën en previews, niet als vervanging voor een nauwkeurige scan.",[10,120,121],{},"Met zo'n snelle rondblik ziet u makkelijker wat werkt aan een personage of object, en waar een andere invoerafbeelding zou helpen.",[83,123,125],{"id":124},"van-392-naar-164-seconden","Van 39,2 naar 16,4 seconden",[10,127,128],{},[90,129],{"alt":130,"src":131},"Volledige Idol-verzoeken op één R9700: standaard geoptimaliseerd 39,22 seconden, Paiton exact 19,45 seconden en Paiton standaard 16,44 seconden. Lager is beter.","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fsame-gpu-nl.webp",[10,133,134],{},[97,135,136],{},"Mediaan van acht afbeeldingen na één opwarmverzoek. Elk verzoek omvat invoerverwerking, generatie en het opslaan van de PLY. Eén R9700, 262.144 Gaussians, 50 stappen, CFG 3,0 en tijdverschuiving 3,0.",[138,139,140,157],"table",{},[141,142,143],"thead",{},[144,145,146,150,154],"tr",{},[147,148,149],"th",{},"Runtime op één R9700",[147,151,153],{"align":152},"right","Volledig verzoek",[147,155,156],{"align":152},"Gemiddelde LPIPS tegenover standaard geoptimaliseerd",[158,159,160,172,186],"tbody",{},[144,161,162,166,169],{},[163,164,165],"td",{},"Standaard-Idol, geoptimaliseerde standaardmodus",[163,167,168],{"align":152},"39,22 s",[163,170,171],{"align":152},"Referentie",[144,173,174,180,183],{},[163,175,176,177],{},"Paiton ",[111,178,179],{},"--exact",[163,181,182],{"align":152},"19,45 s",[163,184,185],{"align":152},"0,0027",[144,187,188,191,196],{},[163,189,190],{},"Paiton standaard",[163,192,193],{"align":152},[23,194,195],{},"16,44 s",[163,197,198],{"align":152},[23,199,200],{},"0,0039",[10,202,203,204,207,208],{},"Dat is ongeveer ",[23,205,206],{},"58% minder wachten",". Downloads, opstarten en het opwarmverzoek vallen buiten de meting. De referentie gebruikt het geoptimaliseerde PyTorch-pad van idol-sdk op ROCm, geen bewust trage basislijn. De GPU behoudt zijn standaardlimiet van 300 W, zonder aanpassingen aan klok, vermogen of ventilator. Die limiet is geen gemeten energieverbruik.",[36,209,210],{},[14,211,45],{"href":40,"ariaDescribedBy":212,"dataFootnoteRef":43,"id":213},[42],"user-content-fnref-bench-2",[10,215,216,217,220,221,223,224],{},"Het meeste werk zit in de denoiser. Paiton vervangt die door een native HIP\u002FC++-runtime voor RDNA4 ",[111,218,219],{},"gfx1201","; de rest blijft draaien op de oorspronkelijke idol-sdk en PyTorch. Matrixproducten blijven fp16. Het standaardprofiel schakelt attention vanaf stap 7 over op int8 QK en fp8 PV; ",[111,222,179],{}," behoudt overal fp16-attention.",[36,225,226],{},[14,227,231],{"href":228,"ariaDescribedBy":229,"dataFootnoteRef":43,"id":230},"#user-content-fn-guide",[42],"user-content-fnref-guide","2",[10,233,234,235,238,239,241,242,244,245,248,249],{},"Hoeveel verandert dat aan het resultaat? Voor de release zijn ",[23,236,237],{},"acht objecten vanuit acht aanzichten per object"," vergeleken met renders van standaard-Idol. De gemiddelde LPIPS was ",[23,240,200],{}," voor het standaardprofiel, ",[23,243,185],{}," voor exact en ",[23,246,247],{},"0,0026"," tussen de geoptimaliseerde en niet-geoptimaliseerde standaardimplementatie. Lager betekent dichter bij elkaar. Het standaardprofiel haalde ook de PSNR-grens voor het slechtst scorende object: 34,63 dB bij een minimum van 33 dB.",[36,250,251],{},[14,252,45],{"href":40,"ariaDescribedBy":253,"dataFootnoteRef":43,"id":254},[42],"user-content-fnref-bench-3",[10,256,257,258,260,261,264],{},"Dit is een afgebakende kwaliteitscontrole, geen controle van elke toekomstige uitvoer. ",[111,259,179],{}," blijft dicht bij de verschillen tussen de standaardimplementaties, maar belooft geen identieke bits of bitreproduceerbaarheid tussen runs. De piek van het ",[23,262,263],{},"door PyTorch gereserveerde GPU-geheugen"," was 9,6 GiB voor beide Paiton-profielen en 10,7 GiB voor standaard-Idol. Dat meet niet het volledige GPU-gebruik en kwalificeert geen kleinere kaarten.",[83,266,268],{"id":267},"waar-de-r9700-uitkomt","Waar de R9700 uitkomt",[10,270,271],{},[90,272],{"alt":273,"src":274},"Gepubliceerde Idol-verzoektijden op verschillende GPUs, met ElioVPs R9700-resultaten apart aangeduid van de metingen van Sperid Labs. Lager is beter.","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fcross-gpu-nl.webp",[10,276,277],{},[97,278,279],{},"R9700-metingen: ElioVP. Andere GPUs: de modelkaart van Sperid Labs. Beide rapporteren 50 stappen, 262.144 Gaussians en medianen over acht afbeeldingen, maar het zijn afzonderlijke metingen, geen gecontroleerde hardwarevergelijking.",[138,281,282,295],{},[141,283,284],{},[144,285,286,289,292],{},[147,287,288],{},"GPU en runtime",[147,290,291],{"align":152},"Verzoek",[147,293,294],{},"Bron",[158,296,297,308,318,330,340,350,359],{},[144,298,299,302,305],{},[163,300,301],{},"H100 80 GB, variant niet vermeld",[163,303,304],{"align":152},"6,7 s",[163,306,307],{},"Sperid Labs",[144,309,310,313,316],{},[163,311,312],{},"RTX 5090",[163,314,315],{"align":152},"14,1 s",[163,317,307],{},[144,319,320,323,327],{},[163,321,322],{},"R9700, Paiton standaard",[163,324,325],{"align":152},[23,326,195],{},[163,328,329],{},"ElioVP",[144,331,332,335,338],{},[163,333,334],{},"L40S",[163,336,337],{"align":152},"16,6 s",[163,339,307],{},[144,341,342,345,348],{},[163,343,344],{},"RTX 4090",[163,346,347],{"align":152},"19,1 s",[163,349,307],{},[144,351,352,355,357],{},[163,353,354],{},"R9700, standaard-Idol",[163,356,168],{"align":152},[163,358,329],{},[144,360,361,364,367],{},[163,362,363],{},"RTX 3080 10 GB",[163,365,366],{"align":152},"46,2 s",[163,368,307],{},[10,370,371,372,375,376],{},"Met deze gepubliceerde tijden haalt de R9700 ongeveer ",[23,373,374],{},"86% van de berekende verzoekdoorvoer van de RTX 5090",". Het cijfer ligt dicht bij dat van de L40S. Zo'n klein verschil tussen afzonderlijke opstellingen levert echter geen betrouwbare rangschikking op.",[36,377,378],{},[14,379,383],{"href":380,"ariaDescribedBy":381,"dataFootnoteRef":43,"id":382},"#user-content-fn-upstream",[42],"user-content-fnref-upstream","3",[385,386,388],"h3",{"id":387},"en-het-hardwarebudget","En het hardwarebudget?",[10,390,391],{},[90,392],{"alt":393,"src":394},"Indicatieve Idol-objecten per uur per 1.000 Amerikaanse dollar aan GPU-aanschaf: R9700 115,3, RTX 3080 109,7, RTX 5090 77,4, RTX 4090 39,3, L40S 28,6 en H100 17,3.","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fgpu-value-nl.webp",[10,396,397],{},[97,398,399],{},"Berekend als 3.600 ÷ verzoektijd in seconden ÷ aangenomen GPU-prijs × 1.000. Een indicatieve vergelijking van alleen de GPU, geen doorvoertest van een uur. Hostkosten, energie en diensten zijn niet inbegrepen.",[138,401,402,415],{},[141,403,404],{},[144,405,406,409,412],{},[147,407,408],{},"GPU",[147,410,411],{"align":152},"Aangenomen prijs, USD",[147,413,414],{"align":152},"Berekende objecten\u002Fuur per $1.000",[158,416,417,429,439,449,459,469],{},[144,418,419,421,424],{},[163,420,322],{},[163,422,423],{"align":152},"$1.900",[163,425,426],{"align":152},[23,427,428],{},"115,3",[144,430,431,433,436],{},[163,432,363],{},[163,434,435],{"align":152},"$710",[163,437,438],{"align":152},"109,7",[144,440,441,443,446],{},[163,442,312],{},[163,444,445],{"align":152},"$3.300",[163,447,448],{"align":152},"77,4",[144,450,451,453,456],{},[163,452,344],{},[163,454,455],{"align":152},"$4.800",[163,457,458],{"align":152},"39,3",[144,460,461,463,466],{},[163,462,334],{},[163,464,465],{"align":152},"$7.570",[163,467,468],{"align":152},"28,6",[144,470,471,474,477],{},[163,472,473],{},"H100 80 GB",[163,475,476],{"align":152},"$31.000",[163,478,479],{"align":152},"17,3",[10,481,482,483,486,487],{},"Dit zijn de prijsaannames uit de gepubliceerde README voor november 2026, ",[23,484,485],{},"geen onafhankelijk bevestigde laagst beschikbare prijzen",". De H100-prijs betreft PCIe-hardware, terwijl de upstreammeting geen variant vermeldt. Vooral die rij is dus een benadering.",[36,488,489],{},[14,490,231],{"href":228,"ariaDescribedBy":491,"dataFootnoteRef":43,"id":492},[42],"user-content-fnref-guide-2",[10,494,495],{},"Bij de aannames van $1.900 en $4.800 kost de R9700 ongeveer 40% van de RTX 4090. Prijzen kunnen de rangschikking veranderen: de RTX 3080 zit hier al dichtbij. Reken met uw eigen offerte, zeker als u al een geschikte GPU hebt.",[83,497,499],{"id":498},"probeer-het-op-uw-r9700","Probeer het op uw R9700",[10,501,502,503,511,512,515,516],{},"U hebt Linux, Docker, een werkend AMD-stuurprogramma en de geteste R9700 nodig. ",[23,504,505,506],{},"Accepteer eerst de ",[14,507,510],{"href":508,"rel":509},"https:\u002F\u002Fhuggingface.co\u002Ftimm\u002Fvit_huge_plus_patch16_dinov3.lvd1689m",[104],"DINOv3-licentie op Hugging Face"," en gebruik daarna een token van dat account. Houd echte tokens privé; ",[111,513,514],{},"hf_..."," hieronder is alleen een voorbeeldwaarde.",[36,517,518],{},[14,519,231],{"href":228,"ariaDescribedBy":520,"dataFootnoteRef":43,"id":521},[42],"user-content-fnref-guide-3",[523,524,528],"pre",{"className":525,"code":526,"language":527,"meta":43,"style":43},"language-bash shiki shiki-themes github-light github-dark","docker pull ghcr.io\u002Feliovp\u002Fpaiton-vllm-plugin:idol-rdna4-v1.0.0\n\nexport HF_TOKEN=hf_...\nmkdir -p \"$HOME\u002Fpaiton-idol\u002Fmodels\" \"$HOME\u002Fpaiton-idol\u002Foutputs\"\ndocker run --rm --device \u002Fdev\u002Fkfd --device \u002Fdev\u002Fdri --group-add \"$(stat -c '%g' \u002Fdev\u002Fkfd)\" \\\n  -e HF_TOKEN -v \"$HOME\u002Fpaiton-idol\u002Fmodels:\u002Fmodels\" -v \"$HOME\u002Fpaiton-idol\u002Foutputs:\u002Foutputs\" -v \"$PWD:\u002Finputs:ro\" \\\n  ghcr.io\u002Feliovp\u002Fpaiton-vllm-plugin:idol-rdna4-v1.0.0 generate \u002Finputs\u002Fphoto.png --mode real\n","bash",[111,529,530,546,553,570,596,636,675],{"__ignoreMap":43},[531,532,535,539,543],"span",{"class":533,"line":534},"line",1,[531,536,538],{"class":537},"sScJk","docker",[531,540,542],{"class":541},"sZZnC"," pull",[531,544,545],{"class":541}," ghcr.io\u002Feliovp\u002Fpaiton-vllm-plugin:idol-rdna4-v1.0.0\n",[531,547,549],{"class":533,"line":548},2,[531,550,552],{"emptyLinePlaceholder":551},true,"\n",[531,554,556,560,564,567],{"class":533,"line":555},3,[531,557,559],{"class":558},"szBVR","export",[531,561,563],{"class":562},"sVt8B"," HF_TOKEN",[531,565,566],{"class":558},"=",[531,568,569],{"class":562},"hf_...\n",[531,571,573,576,580,583,586,589,591,593],{"class":533,"line":572},4,[531,574,575],{"class":537},"mkdir",[531,577,579],{"class":578},"sj4cs"," -p",[531,581,582],{"class":541}," \"",[531,584,585],{"class":562},"$HOME",[531,587,588],{"class":541},"\u002Fpaiton-idol\u002Fmodels\"",[531,590,582],{"class":541},[531,592,585],{"class":562},[531,594,595],{"class":541},"\u002Fpaiton-idol\u002Foutputs\"\n",[531,597,599,601,604,607,610,613,615,618,621,624,627,630,633],{"class":533,"line":598},5,[531,600,538],{"class":537},[531,602,603],{"class":541}," run",[531,605,606],{"class":578}," --rm",[531,608,609],{"class":578}," --device",[531,611,612],{"class":541}," \u002Fdev\u002Fkfd",[531,614,609],{"class":578},[531,616,617],{"class":541}," \u002Fdev\u002Fdri",[531,619,620],{"class":578}," --group-add",[531,622,623],{"class":541}," \"$(",[531,625,626],{"class":578},"stat",[531,628,629],{"class":578}," -c",[531,631,632],{"class":541}," '%g' \u002Fdev\u002Fkfd)\"",[531,634,635],{"class":578}," \\\n",[531,637,639,642,644,647,649,651,654,656,658,660,663,665,667,670,673],{"class":533,"line":638},6,[531,640,641],{"class":578},"  -e",[531,643,563],{"class":541},[531,645,646],{"class":578}," -v",[531,648,582],{"class":541},[531,650,585],{"class":562},[531,652,653],{"class":541},"\u002Fpaiton-idol\u002Fmodels:\u002Fmodels\"",[531,655,646],{"class":578},[531,657,582],{"class":541},[531,659,585],{"class":562},[531,661,662],{"class":541},"\u002Fpaiton-idol\u002Foutputs:\u002Foutputs\"",[531,664,646],{"class":578},[531,666,582],{"class":541},[531,668,669],{"class":562},"$PWD",[531,671,672],{"class":541},":\u002Finputs:ro\"",[531,674,635],{"class":578},[531,676,678,681,684,687,690],{"class":533,"line":677},7,[531,679,680],{"class":541},"  ghcr.io\u002Feliovp\u002Fpaiton-vllm-plugin:idol-rdna4-v1.0.0",[531,682,683],{"class":541}," generate",[531,685,686],{"class":541}," \u002Finputs\u002Fphoto.png",[531,688,689],{"class":578}," --mode",[531,691,692],{"class":541}," real\n",[10,694,695,696,699,700,703,704,706],{},"Plaats ",[111,697,698],{},"photo.png"," in uw huidige map. Gebruik ",[111,701,702],{},"--mode synthetic"," voor renders of gegenereerde afbeeldingen, zoals in de benchmark. Voeg ",[111,705,179],{}," toe om fp16-attention te behouden. De standaardinstellingen zijn 50 stappen, CFG 3,0, seed 42 en 262.144 Gaussians.",[10,708,709,710,713,714,716,717,720,721,724],{},"De eerste run downloadt de gewichten; volgende runs hergebruiken de modelmap. In ",[111,711,712],{},"$HOME\u002Fpaiton-idol\u002Foutputs"," verschijnen een ",[111,715,113],{},", de voorbereide ",[111,718,719],{},"_input.png"," en een ",[111,722,723],{},"_camera.json",". Voor een vastgezette containerreferentie: op 6 november 2026 verwees de openbare tag naar deze registry-digest:",[523,726,731],{"className":727,"code":729,"language":730,"meta":43},[728],"language-text","ghcr.io\u002Feliovp\u002Fpaiton-vllm-plugin@sha256:7939b4e0bcacf49bcce51d71892915eda802ad2f3724ef61e06e1bc7a77b4e5d\n","text",[111,732,729],{"__ignoreMap":43},[10,734,735,736,741],{},"Vervang de tag in beide Docker-commando's door die referentie om het image vast te zetten. De ",[14,737,740],{"href":738,"rel":739},"https:\u002F\u002Fgithub.com\u002FEliovp-BV\u002Fpaiton-vllm-plugin\u002Fblob\u002F9ff80088fb0063fe0dc140200c21ebc262594aac\u002Fmodels\u002FIdol\u002FREADME.md#run",[104],"openbare releasehandleiding"," beschrijft het starten en downloaden.",[10,743,744,745],{},"De pakketcode en uitgebrachte native runtime vallen onder Apache-2.0; de compiler van Paiton blijft gesloten. Idol en de FLUX.2 klein-VAE hebben Apache-2.0-voorwaarden, terwijl DINOv3 zijn eigen licentie behoudt. De gewichten worden afzonderlijk gedownload.",[36,746,747],{},[14,748,752],{"href":749,"ariaDescribedBy":750,"dataFootnoteRef":43,"id":751},"#user-content-fn-licenses",[42],"user-content-fnref-licenses","4",[10,754,755,756,760,761,765,766,770],{},"Weer een nuttige toepassing voor hetzelfde lokale werkstation. Ontdek ",[14,757,759],{"href":758},"\u002Fnl\u002Fproducts\u002Fpaiton","Paiton",", onze ",[14,762,764],{"href":763},"\u002Fnl\u002Fblog\u002Fpaiton-qwen-image-21-radeon-ai-pro-r9700","lokale beeldgeneratie"," of ",[14,767,769],{"href":768},"\u002Fnl\u002Fproducts\u002Fsovereign-ai","Sovereign AI"," om deze workflows op uw eigen infrastructuur te houden.",[772,773,776,781],"section",{"className":774,"dataFootnotes":43},[775],"footnotes",[83,777,780],{"className":778,"id":42},[779],"sr-only","Footnotes",[782,783,784,815,841,854],"ol",{},[58,785,787,792,793,800,801,800,808],{"id":786},"user-content-fn-bench",[14,788,791],{"href":789,"rel":790},"https:\u002F\u002Fgithub.com\u002FEliovp-BV\u002Fpaiton-vllm-plugin\u002Fblob\u002F9ff80088fb0063fe0dc140200c21ebc262594aac\u002Fmodels\u002FIdol\u002FBENCHMARKS.md",[104],"Idol-releasebenchmark en kwaliteitscontrole",". ",[14,794,799],{"href":795,"ariaLabel":796,"className":797,"dataFootnoteBackref":43},"#user-content-fnref-bench","Back to reference 1",[798],"data-footnote-backref","↩"," ",[14,802,799,806],{"href":803,"ariaLabel":804,"className":805,"dataFootnoteBackref":43},"#user-content-fnref-bench-2","Back to reference 1-2",[798],[36,807,231],{},[14,809,799,813],{"href":810,"ariaLabel":811,"className":812,"dataFootnoteBackref":43},"#user-content-fnref-bench-3","Back to reference 1-3",[798],[36,814,383],{},[58,816,818,792,822,800,827,800,834],{"id":817},"user-content-fn-guide",[14,819,821],{"href":102,"rel":820},[104],"Paiton Idol-installatie, runtimedetails en gepubliceerde prijsaannames",[14,823,799],{"href":824,"ariaLabel":825,"className":826,"dataFootnoteBackref":43},"#user-content-fnref-guide","Back to reference 2",[798],[14,828,799,832],{"href":829,"ariaLabel":830,"className":831,"dataFootnoteBackref":43},"#user-content-fnref-guide-2","Back to reference 2-2",[798],[36,833,231],{},[14,835,799,839],{"href":836,"ariaLabel":837,"className":838,"dataFootnoteBackref":43},"#user-content-fnref-guide-3","Back to reference 2-3",[798],[36,840,383],{},[58,842,844,792,849],{"id":843},"user-content-fn-upstream",[14,845,848],{"href":846,"rel":847},"https:\u002F\u002Fhuggingface.co\u002Fsperidlabs\u002Fidol\u002Fblob\u002F2731c2291fc65a797c5cb627a5e5526fc1490902\u002FREADME.md#across-gpus",[104],"Sperid Labs Idol-modelkaart, nagekeken revisie",[14,850,799],{"href":851,"ariaLabel":852,"className":853,"dataFootnoteBackref":43},"#user-content-fnref-upstream","Back to reference 3",[798],[58,855,857,792,862],{"id":856},"user-content-fn-licenses",[14,858,861],{"href":859,"rel":860},"https:\u002F\u002Fgithub.com\u002FEliovp-BV\u002Fpaiton-vllm-plugin\u002Fblob\u002F9ff80088fb0063fe0dc140200c21ebc262594aac\u002Fmodels\u002FIdol\u002FTHIRD_PARTY_NOTICES.md",[104],"Runtimegrenzen en licenties van derden",[14,863,799],{"href":864,"ariaLabel":865,"className":866,"dataFootnoteBackref":43},"#user-content-fnref-licenses","Back to reference 4",[798],[868,869,870],"style",{},"html pre.shiki code .sScJk, html code.shiki .sScJk{--shiki-default:#6F42C1;--shiki-dark:#B392F0}html pre.shiki code .sZZnC, html code.shiki .sZZnC{--shiki-default:#032F62;--shiki-dark:#9ECBFF}html pre.shiki code .szBVR, html code.shiki .szBVR{--shiki-default:#D73A49;--shiki-dark:#F97583}html pre.shiki code .sVt8B, html code.shiki .sVt8B{--shiki-default:#24292E;--shiki-dark:#E1E4E8}html pre.shiki code .sj4cs, html code.shiki .sj4cs{--shiki-default:#005CC5;--shiki-dark:#79B8FF}html .default .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .shiki span {color: var(--shiki-default);background: var(--shiki-default-bg);font-style: var(--shiki-default-font-style);font-weight: var(--shiki-default-font-weight);text-decoration: var(--shiki-default-text-decoration);}html .dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}html.dark .shiki span {color: var(--shiki-dark);background: var(--shiki-dark-bg);font-style: var(--shiki-dark-font-style);font-weight: var(--shiki-dark-font-weight);text-decoration: var(--shiki-dark-text-decoration);}",{"title":43,"searchDepth":548,"depth":548,"links":872},[873,874,875,878,879],{"id":85,"depth":548,"text":86},{"id":124,"depth":548,"text":125},{"id":267,"depth":548,"text":268,"children":876},[877],{"id":387,"depth":555,"text":388},{"id":498,"depth":548,"text":499},{"id":42,"depth":548,"text":780},[759,881,882,883],"AMD Radeon","Lokale AI","3D-generatie","2026-11-06T09:00:00Z","Idol maakt een 3D Gaussian splat in 16,44 seconden op één Radeon AI PRO R9700 met Paiton, tegenover 39,22 seconden voor standaard-Idol op dezelfde GPU.","md","De R9700 doet ook 3D. Idol in 16,4 seconden.","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fhero-nl.webp",{},"https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-idol-radeon-ai-pro-r9700","\u002Fblog\u002Fpaiton-idol-radeon-ai-pro-r9700",{"title":5,"description":885},"paiton-idol-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-idol\u002Fsocial-nl.webp","blog\u002Fpaiton-idol-radeon-ai-pro-r9700",null,"12e_WU6K7mBxU1Dp_jyz-LRt8FzeVHAU0pifEhppDSE",[899,901,914,924,933,948,957,966,975,985,997,1007,1019,1029,1042,1051,1063,1093,1105,1127,1145,1164,1183,1200,1217,1228,1244,1259,1271,1280,1288,1303,1315,1326,1337,1347,1360,1370,1383,1394,1404,1415,1424,1436,1446,1455],{"path":891,"title":5,"description":885,"date":884,"slug":893,"image":888,"originalUrl":890,"categories":900},[759,881,882,883],{"path":902,"title":903,"description":904,"date":905,"slug":906,"image":907,"originalUrl":908,"categories":909},"\u002Fblog\u002Fpaiton-deepseek-v41-flash-amd-instinct-mi350p","Deepseek V4.1 Flash op één MI350P","DeepSeek V4.1 Flash op één MI350P met 144 GB: 77,9 tokens\u002Fs bij één verzoek en 1.107,6 in een batch van 512, met Paiton en expertberekeningen op CPU.","2026-11-03T12:00:00Z","paiton-deepseek-v41-flash-amd-instinct-mi350p","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-deepseek-v41\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-deepseek-v41-flash-amd-instinct-mi350p",[759,910,911,912,913],"AMD Instinct","AI-inferentie","DeepSeek","Kwantisatie",{"path":915,"title":916,"description":917,"date":918,"slug":919,"image":920,"originalUrl":921,"categories":922},"\u002Fblog\u002Fpaiton-qwen38-flash-next-amd-instinct-mi350p","Qwen3.8 Flash Next op MI350P: dubbel zo snel | Paiton","Qwen3.8 Flash Next op één AMD MI350P: 2,24× de decodesnelheid van de geteste vLLM-configuratie bij één verzoek en circa 2,2 miljoen gedeelde KV-cachetokens.","2026-10-30T12:00:00Z","paiton-qwen38-flash-next-amd-instinct-mi350p","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-mi350p\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-flash-next-amd-instinct-mi350p",[759,910,911,923,913],"Qwen",{"path":925,"title":926,"description":927,"date":928,"slug":929,"image":930,"originalUrl":931,"categories":932},"\u002Fblog\u002Fpaiton-qwen38-3bit-accuracy-prefix-caching-mtp","Qwen3.8 Flash Next: 3-bit dicht bij 16-bit nauwkeurigheid, met MTP-prefixcaching","Paitons nieuwe 3-bit Qwen3.8-gewichten benaderen 16-bit taakscores op twee R9700-kaarten. Native MTP-prefixcaching hervat een prompt van 200K in 0,68 seconde. Resultaten, grenzen en installatie.","2026-10-24T18:45:00Z","paiton-qwen38-3bit-accuracy-prefix-caching-mtp","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen38-3bit-mtp\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-3bit-accuracy-prefix-caching-mtp",[759,881,882,923,913],{"path":934,"title":935,"description":936,"date":937,"slug":938,"image":939,"originalUrl":940,"categories":941},"\u002Fblog\u002Fnvidia-dsx-pod-vera-rubin-modular-data-centers","Nvidia confirms the modular DC direction","NVIDIA DSX POD brengt Vera Rubin AI-fabrieken naar modulaire datacenters. ElioVP bespreekt terreinvoorbereiding, koeling en de aanpak achter ModFlex.","2026-10-21T13:00:00Z","nvidia-dsx-pod-vera-rubin-modular-data-centers","\u002Fasset\u002Fimages\u002Fblog\u002Fnvidia-dsx-pod-modflex\u002Fhero.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fnvidia-dsx-pod-vera-rubin-modular-data-centers",[942,943,944,945,946,947],"Modular DC","AI-infrastructuur","ModFlex","NVIDIA","Vera Rubin","Vloeistofkoeling",{"path":949,"title":950,"description":951,"date":952,"slug":953,"image":954,"originalUrl":955,"categories":956},"\u002Fblog\u002Fpaiton-qwen38-flash-next-4bit-radeon-ai-pro-r9700","Qwen3.8 Flash Next in 4-bit op 2 × Radeon R9700: getest tegenover 16-bit","Paiton draait Qwen3.8 Flash Next in 4-bit op twee R9700-kaarten: 96,8% GSM8K, 96,3% HumanEval, 189,8 tokens\u002Fs en een optionele RAM-prefixcache. Kwaliteit, snelheid en installatie.","2026-10-12T15:00:00Z","paiton-qwen38-flash-next-4bit-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen38-flash-next-4bit\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-flash-next-4bit-radeon-ai-pro-r9700",[759,881,882,923,913],{"path":958,"title":959,"description":960,"date":961,"slug":962,"image":963,"originalUrl":964,"categories":965},"\u002Fblog\u002Fpaiton-qwen38-flash-next-radeon-ai-pro-r9700","Qwen3.8 Flash Next op 2 × Radeon AI PRO R9700: 216 tokens\u002Fs, 200K-context","Paiton draait Qwen3.8 Flash Next met 3-bit experts op twee R9700-kaarten: 216,3 tokens\u002Fs, 565,1 tokens\u002Fs samen en een aparte 200K-modus. Metingen, kwaliteit en installatie.","2026-10-10T10:30:00Z","paiton-qwen38-flash-next-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen38-flash-next\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-flash-next-radeon-ai-pro-r9700",[759,881,882,923,913],{"path":967,"title":968,"description":969,"date":970,"slug":971,"image":972,"originalUrl":973,"categories":974},"\u002Fblog\u002Fpaiton-qwen38-w3a4-radeon-ai-pro-r9700","Qwen3.8 27B op 1 × Radeon AI PRO R9700: 3-bit gewichten, 20% snellere decode","Qwen3.8 27B op één R9700: 19,9% snellere gewogen decode, een nieuwe 4-bit cache, 200K-context en optionele beeldinvoer. Metingen en Paiton-installatie.","2026-09-27T09:00:00Z","paiton-qwen38-w3a4-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen38-w3a4\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen38-w3a4-radeon-ai-pro-r9700",[759,881,882,923,913],{"path":976,"title":977,"description":978,"date":979,"slug":980,"image":981,"originalUrl":982,"categories":983},"\u002Fblog\u002Fpaiton-qwen-image-21-radeon-ai-pro-r9700","Qwen-Image 2.1 op 1 × Radeon AI PRO R9700: 2048×2048-beelden in 103 seconden","Genereer lokaal Qwen-Image 2.1-beelden van 2048×2048 pixels met 1 × Radeon AI PRO R9700. De Paiton v1.0.2-container mat 103,29 seconden per opgewarmd verzoek, inclusief de PNG-respons.","2026-09-23T09:00:00Z","paiton-qwen-image-21-radeon-ai-pro-r9700","\u002Fasset\u002Fimages\u002Fblog\u002Fpaiton-qwen-image-21\u002Fhero-nl.webp","https:\u002F\u002Feliovp.com\u002Fblog\u002Fpaiton-qwen-image-21-radeon-ai-pro-r9700",[759,881,882,984,923],"Beeldgeneratie",{"path":986,"title":987,"description":988,"date":989,"slug":990,"image":991,"originalUrl":992,"categories":993},"\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",[759,881,882,911,994,995,996],"Inferentie-optimalisatie","Grote taalmodellen","vLLM",{"path":998,"title":999,"description":1000,"date":1001,"slug":1002,"image":1003,"originalUrl":1004,"categories":1005},"\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",[759,881,882,1006,996],"GGUF",{"path":1008,"title":1009,"description":1010,"date":1011,"slug":1012,"image":1013,"originalUrl":1014,"categories":1015},"\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",[759,881,882,1016,1017,1018],"Videogeneratie","MiniMax H3","ComfyUI",{"path":1020,"title":1021,"description":1022,"date":1023,"slug":1024,"image":1025,"originalUrl":1026,"categories":1027},"\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",[759,881,882,984,1028,1018],"FLUX",{"path":1030,"title":1031,"description":1032,"date":1033,"slug":1034,"image":1035,"originalUrl":1036,"categories":1037},"\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",[759,1038,881,911,1039,1040,994,995,996,1041],"Kunstmatige intelligentie","GPU-prestaties","Inferentielatentie","Kostenefficiëntie",{"path":1043,"title":1044,"description":1045,"date":1046,"slug":1047,"image":1048,"originalUrl":1049,"categories":1050},"\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",[759,1038,881,911,1039,1040,994,995,996,1041],{"path":1052,"title":1053,"description":1054,"date":1055,"slug":1056,"image":1057,"originalUrl":896,"categories":1058},"\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",[1059,943,1060,944,1061,1062],"Alle","Datacenters","HPC","AMD Helios",{"path":1064,"title":1065,"description":1066,"date":1067,"slug":1068,"image":1069,"originalUrl":1070,"categories":1071},"\u002Fblog\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x","Wan2.2-videogeneratie: Paiton op AMD MI355X","Vergelijk Wan2.2-videogeneratie op AMD MI355X met Paiton en NVIDIA B200 via Diffusers. Lees hoe we diffusiemodellen optimaliseren.","2026-06-10T14:04:04","paiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x","\u002Fasset\u002Fimages\u002Fblog\u002Fimported\u002Fpaitonwan2.webp","https:\u002F\u002Feliovp.com\u002Fpaiton-returns-to-its-diffusion-roots-optimizing-wan2-2-t2v-a14b-on-amd-mi355x\u002F",[1059,1038,759,1072,1073,1074,1075,1076,1077,1078,1079,1080,408,1081,1082,1083,1084,945,1085,1086,759,1087,1088,1089,1090,1091,1092],"14B","AMD","B200","Benchmarks","Blackwell","Compute","Diffusie","Eliovp","Generatieve AI","Hardware","Inferentie","Instinct","MI355x","On-premises","Optimalisatie","Soevereine AI","T2V","Tekst-naar-video","Tuning","Video-generatie","Wan2.2",{"path":1094,"title":1095,"description":1096,"date":1097,"slug":1098,"image":1099,"originalUrl":1100,"categories":1101},"\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",[1059,1038,1102,1103,1073,1104,1102,945],"Modulaire DC","Niet gecategoriseerd","De Tijd",{"path":1106,"title":1107,"description":1108,"date":1109,"slug":1110,"image":1111,"originalUrl":1112,"categories":1113},"\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",[1059,1038,1114,1103,1115,1116,1117,1118,1119,1120,1121,1122,1079,1123,1080,1124,1125,1126],"Trending","AI Act","Antropomorfisme","AVG","Benelux","ChatGPT","Cyberbeveiliging","Databeheer","Gegevensbeveiliging","GDPR","Microsoft Copilot","Privacy","Shadow AI",{"path":1128,"title":1129,"description":1130,"date":1131,"slug":1132,"image":1133,"originalUrl":1134,"categories":1135},"\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",[1059,1136,1137,1138,1120,1139,1140,1141,1123,1142,1143,1144,1125],"AWS","Belgian Mobile ID","CLOUD Act","Datasoevereiniteit","Digitale identiteit","eIDAS","itsme","Liberty Global","MyGov.be",{"path":1146,"title":1147,"description":1148,"date":1149,"slug":1150,"image":1151,"originalUrl":1152,"categories":1153},"\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",[1059,1038,1154,1114,1155,1156,1157,1158,1159,1160,1161,1162,1087,1163],"Oplossingen","Agentic AI","AI-techniek","AI-strategie","Autonome agenten","Bedrijfs-AI","Lokale LLM","Modelverfijning","AI op locatie","VRAM-optimalisatie",{"path":1165,"title":1166,"description":1167,"date":1168,"slug":1169,"image":1170,"originalUrl":1171,"categories":1172},"\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",[1059,1038,1114,1173,1174,1175,1176,1177,1178,1179,1180,1181,1182],"AI Infrastructure","AI Neocloud","Circulaire financiering","GPU Cloud","Beleggingsrisico's","Opstartwaardering","Synthetische bubbel","Technische analyse","Vaporware","Durfkapitaal",{"path":1184,"title":1185,"description":1186,"date":1187,"slug":1188,"image":1189,"originalUrl":1190,"categories":1191},"\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",[1059,1102,1103,1192,943,1193,1194,947,1195,1196,1197,1198,1199],"150 kW-rack","DLC","Hoge dichtheid","Modulair datacenter","NVIDIA Blackwell Ultra","NVIDIA GB300","NVL72","Snelle implementatie",{"path":1201,"title":1202,"description":1203,"date":1204,"slug":1205,"image":1206,"originalUrl":1207,"categories":1208},"\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",[1059,1038,759,1103,1209,1038,1210,1211,1212,1213,1214,1215,759,1216],"AMD MI300X","CUDA-vertaling","FP8","GPU-optimalisatie","High-performance computing","HIP","Kerneltuning","ROCm",{"path":1218,"title":1219,"description":1220,"date":1221,"slug":1222,"image":1223,"originalUrl":1224,"categories":1225},"\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",[1059,1038,759,911,910,1209,1041,1226,994,1215,759,1227,996],"Hoge throughput","SGLang",{"path":1229,"title":1230,"description":1231,"date":1232,"slug":1233,"image":1234,"originalUrl":1235,"categories":1236},"\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",[1059,1038,759,1237,1209,1238,994,1239,1240,1241,1242,759,1243],"AI-benchmarks","Kosten per token","Mixture of Experts","MoE","NVIDIA B200","NVIDIA H200","Qwen3",{"path":1245,"title":1246,"description":1247,"date":1248,"slug":1249,"image":1250,"originalUrl":1251,"categories":1252},"\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",[1059,1038,1154,1103,1155,1253,1254,1255,1256,1160,1162,1087,1257,1258],"Schadedetectie","Documentverwerking","E-mailautomatisering","Factuurextractie","Ticketautomatisering","Workflowautomatisering",{"path":1260,"title":1261,"description":1262,"date":1263,"slug":1264,"image":1265,"originalUrl":1266,"categories":1267},"\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",[1059,1038,759,1103,1268,1073,1074,1269,1270,1084,945,759,996],"AI","H200","MI300X",{"path":1272,"title":1273,"description":1274,"date":1275,"slug":1276,"image":1277,"originalUrl":1278,"categories":1279},"\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",[1059,1038,1154,1155,1253,1254,1255,1256,1160,1162,1087,1257,1258],{"path":1281,"title":1282,"description":1283,"date":1284,"slug":1285,"image":43,"originalUrl":1286,"categories":1287},"\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",[1059,1038,759],{"path":1289,"title":1290,"description":1291,"date":1292,"slug":1293,"image":1294,"originalUrl":1295,"categories":1296},"\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",[1059,1038,759,1103,911,1209,1297,1211,1298,1299,1300,759,1301,1302],"Koude start","Grafiekcompilatie","Llama 3.1 405B","LLM-optimalisatie","Opstartlatentie","Tensor-parallellisme",{"path":1304,"title":1305,"description":1306,"date":1307,"slug":1308,"image":1309,"originalUrl":1310,"categories":1311},"\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",[1059,1038,759,1103,1209,1312,1159,1080,1039,1040,995,1299,1313,1314],"Koude startoptimalisatie","Model serving","vLLM-optimalisatie",{"path":1316,"title":1317,"description":1318,"date":1319,"slug":1320,"image":1321,"originalUrl":1322,"categories":1323},"\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",[1059,1038,759,1154,1103,1073,1270,945,1324,1325],"RX7900XTX","tenstorrent",{"path":1327,"title":1328,"description":1329,"date":1330,"slug":1331,"image":1332,"originalUrl":1333,"categories":1334},"\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",[1059,1038,1102,1154,1074,1269,1335,945,1336],"MI325X","P&L-calculator",{"path":1338,"title":1339,"description":1340,"date":1341,"slug":1342,"image":1343,"originalUrl":1344,"categories":1345},"\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",[1059,1038,759,1268,1073,1269,1346,945,759,996],"MI300",{"path":1348,"title":1349,"description":1350,"date":1351,"slug":1352,"image":1353,"originalUrl":1354,"categories":1355},"\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",[1059,1102,1356,1173,1194,1061,947,1195,1357,1358,1198,1359],"1-2MW datacenter","NVIDIA Blackwell","NVIDIA NVL","Precisiekoeling",{"path":1361,"title":1362,"description":1363,"date":1364,"slug":1365,"image":1366,"originalUrl":1367,"categories":1368},"\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",[1059,1038,1154,1103,1268,1073,1369],"Zorg",{"path":1371,"title":1372,"description":1373,"date":1374,"slug":1375,"image":1376,"originalUrl":1377,"categories":1378},"\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",[1059,1114,1268,1073,1379,1380,1381,1382],"Invoer","Taiwan","Invoerheffingen","Trump",{"path":1384,"title":1385,"description":1386,"date":1387,"slug":1388,"image":1389,"originalUrl":1390,"categories":1391},"\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",[1059,1038,1154,1268,1392,1393],"AI-agenten","ERP",{"path":1395,"title":1396,"description":1397,"date":1398,"slug":1399,"image":1400,"originalUrl":1401,"categories":1402},"\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",[1059,1038,1114,1403,1073,408,945],"AI-nieuws",{"path":1405,"title":1406,"description":1407,"date":1408,"slug":1409,"image":1410,"originalUrl":1411,"categories":1412},"\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",[1059,1038,759,1268,1073,1413,1414,1270,759],"benchmark","LLM",{"path":1416,"title":1417,"description":1418,"date":1419,"slug":1420,"image":1421,"originalUrl":1422,"categories":1423},"\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",[1059,1038,759],{"path":1425,"title":1426,"description":1427,"date":1428,"slug":1429,"image":1430,"originalUrl":1431,"categories":1432},"\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",[1059,1073,1433,1434,1435],"Jim Greene","Podcast","Tech Talk",{"path":1437,"title":1438,"description":1439,"date":1440,"slug":1441,"image":1442,"originalUrl":1443,"categories":1444},"\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",[1059,1038,759,1073,912,1445,1269,1270,1335,759,996],"H100",{"path":1447,"title":1448,"description":1449,"date":1450,"slug":1451,"image":1452,"originalUrl":1453,"categories":1454},"\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",[1059,1038,759,1073,912,1445,1269,1270,1335,759,996],{"path":1456,"title":1457,"description":1458,"date":1459,"slug":1460,"image":1461,"originalUrl":1462,"categories":1463},"\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",[1059,1038,759,1073,1445,1269,1270,1335,759,996],1793962345768]