{"id":12156019278164,"title":"Módulo NVIDIA Jetson T5000","handle":"modulo-nvidia-jetson-t5000","description":"\u003cdiv id=\"buildify\" class=\"buildify buildify-12156019278164\"\u003e\u003cdiv class=\"bdf-innr\"\u003e \u003cdiv class=\"bdf-s-wrp\"\u003e \u003csection data-id=\"ywqzfdc\" class=\"bdf-e bdf-e-ywqzfdc bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-top-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"awuaawd\" class=\"bdf-e bdf-e-awuaawd bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"lfuzqeo\" class=\"bdf-e bdf-e-lfuzqeo bdf-w bdf-w-heading\" data-el=\"heading.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003ch2 class=\"bdf-h-t bdf-sz-medium\"\u003eModulo NVIDIA Jetson T5000\u003c\/h2\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"zgedg2q\" class=\"bdf-e bdf-e-zgedg2q bdf-w bdf-w-heading\" data-el=\"heading.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003ch2 class=\"bdf-h-t bdf-sz-small\"\u003e900-13834-0080-000\u003c\/h2\u003e \u003c\/div\u003e \u003c\/div\u003e \u003csection data-id=\"dzrjrze\" class=\"bdf-e bdf-e-dzrjrze bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"jachcmk\" class=\"bdf-e bdf-e-jachcmk bdf-c bdf-c-s-100 bdf-inner-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"ulccafn\" class=\"bdf-e bdf-e-ulccafn bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\u003cul\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eMódulo NVIDIA Jetson T5000\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003e2070 FP4 TFLOPS Computação de IA powered by Blackwell GPU\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eCPU ARM Neoverse de 4 núcleos\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eMemória LPDDR5X de 128 GB\u003c\/span\u003e\u003c\/li\u003e\n\u003c\/ul\u003e\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003cdiv data-id=\"vegfcxn\" class=\"bdf-e bdf-e-vegfcxn bdf-tbs-v-horizontal bdf-w bdf-w-tabs\" data-el=\"tabs.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-tbs bdf-rcv-tb-w\" role=\"tablist\"\u003e \u003cdiv class=\"bdf-tbs-wrp\" role=\"tab\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"1\"\u003eVista Rápida\u003c\/div\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"2\"\u003eCaracterísticas\u003c\/div\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"3\"\u003eDownloads\u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tbs-cnt-wrp bdf-rcv-tb-wrp\" role=\"tabpanel\"\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"1\"\u003eVista Rápida\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-kfp0noq\" data-tab=\"1\"\u003e \u003csection data-id=\"kxfflkj\" class=\"bdf-e bdf-e-kxfflkj bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"qkyonbf\" class=\"bdf-e bdf-e-qkyonbf bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"heduntc\" class=\"bdf-e bdf-e-heduntc bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003e\u003cstrong\u003eA Plataforma Definitiva para AI Física e Robótica\u003c\/strong\u003e\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO NVIDIA® Jetson T5000™ oferece o mais alto nível de desempenho para AI física e robótica. Você obtém até 2070 FP4 TFLOPS de computação em AI e 128 GB de memória com potência configurável entre 40 W e 130 W. Comparado ao AGX Orin™, ele fornece mais de 7.5x mais computação em AI do que o NVIDIA AGX Orin™, com 3.5x melhor eficiência energética.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eEste sistema-em-módulo apresenta uma GPU com arquitetura NVIDIA Blackwell, um motor de transformador e suporte para GPU de Multi-Instância (MIG) para executar facilmente os mais recentes modelos generativos de AI. O NVIDIA Jetson Thor acelera aplicações multi-sensor em tempo real e de baixa latência com uma CPU Arm® Neoverse®-V3AE de 14 núcleos, 4x rede 25 GbE e extensas opções de E\/S para fusão de sensores. Também inclui um conjunto de aceleradores, como um Acelerador de Visão Programável (PVA) de terceira geração, codificadores e decodificadores duplos, um acelerador de fluxo óptico, e mais.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO Jetson Thor pertence a uma nova classe de computadores robóticos, projetados do zero para alimentar aplicações AI físicas de próxima geração. Ele suporta uma ampla gama de modelos generativos de AI, desde modelos de Ação de Linguagem Visual (VLA) como o NVIDIA Isaac™ GR00T para humanos até todos os modelos LLMs e VLMs populares. Para oferecer uma experiência perfeita de nuvem à borda, o Jetson Thor executa a pilha de software AI da NVIDIA para aplicações de AI física, incluindo NVIDIA Isaac para robótica, NVIDIA Metropolis para AI visual agente e NVIDIA Holoscan para processamento de sensores. Você também pode construir agentes de AI na borda usando fluxos de trabalho de AI agente da NVIDIA, como Pesquisa e Resumo de Vídeo (VSS).\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eNosso ecossistema de parceiros oferece todas as placas de transporte, serviços de design, câmaras e outros sensores de que você precisa, bem como software adicional de AI e sistema. Isso permite acelerar o desenvolvimento de soluções em indústrias que vão desde robótica e espaços inteligentes, até retalho, industrial, médica e muito mais.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO módulo Jetson Thor oferece desempenho pronto para produção, computação massiva de AI e capacidades de sensores para aplicações de AI física em um formato compacto. Isso torna-o a plataforma ideal para desenvolvedores que procuram desbloquear novas possibilidades para robótica humanoide e outras aplicações de AI física.\u003c\/span\u003e\u003c\/p\u003e\n\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"2\"\u003eCaracterísticas\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-1jpf9wf\" data-tab=\"2\"\u003e \u003csection data-id=\"19sojk8\" class=\"bdf-e bdf-e-19sojk8 bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"7fzd5zq\" class=\"bdf-e bdf-e-7fzd5zq bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"27otq25\" class=\"bdf-e bdf-e-27otq25 bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\n\u003cdiv class=\"nv-title text h--medium aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-title-6f099a7d2d\" class=\"general-container-text \"\u003e\u003cdiv class=\"text-center lap-text-center tab-text-center mob-text-center\"\u003e\u003ch2 class=\"title \"\u003eNVIDIA Jetson Thor Series\u003c\/h2\u003e\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv class=\"nv-separator separator aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-separator-60a56b94ec\" class=\"cmp-separator\"\u003e \u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv class=\"nv-text text aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-text-1228782b3d\" class=\"general-container-text \"\u003e\u003cdiv class=\"text-left lap-text-left tab-text-left mob-text-left\"\u003e\u003cdiv class=\"description \"\u003e\n\u003cdiv id=\"jetson-prod-module-table-topscroll\"\u003e\u003cdiv class=\"scroll-div1\"\u003e \u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv id=\"jetson-prod-module-table\" class=\"jetson-specs-table scrolling expanded\"\u003e\u003cdiv class=\"scroll-div2\"\u003e\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" align=\"center\"\u003e\u003ctbody\u003e\n\u003ctr class=\"head border\"\u003e\n\u003ctd class=\"titlehead tableCLdata1\" width=\"10%\"\u003e \u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson AGX Thor Developer Kit\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson T5000\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson T4000\u003c\/strong\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eAI Performance\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2070 TFLOPS (FP4—Sparse)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1200 TFLOPS (FP4—Sparse)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eGPU\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2560-core NVIDIA Blackwell architecture GPU with fifth-gen Tensor Cores\u003cbr\u003eMulti-Instance GPU (MIG) with 10 TPCs\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1536-core NVIDIA Blackwell architecture GPU with fifth-gen Tensor Cores\u003cbr\u003eMulti-Instance GPU (MIG) with 6 TPCs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eGPU Max Frequency\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e1.57 GHz\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1.53 GHz\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCPU\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e14-core Arm® Neoverse®-V3AE 64-bit CPU\u003cbr\u003e1 MB L2 cache per core\u003cbr\u003e16 MB shared system L3 cache\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e12-core Arm® Neoverse®-V3AE 64-bit CPU\u003cbr\u003e1 MB L2 cache per core\u003cbr\u003e16 MB shared system L3 cache\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCPU Max Frequency\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"3\"\u003e2.6 GHz\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003e\u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/pva\" target=\"_blank\" rel=\"noopener\"\u003eVision Accelerator\u003c\/a\u003e\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"3\"\u003e1x PVA v3\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eMemory\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e128 GB 256-bit LPDDR5X\u003cbr\u003e273 GB\/s\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e64 GB 256-bit LPDDR5X\u003cbr\u003e273 GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eStorage\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1 TB NVMe M.2 Key M Slot\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eSupports NVMe through PCIe\u003cbr\u003eSupports SSD through USB3.2\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eVideo Encode\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2X NVENCODE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1X NVENCODE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eVideo Decode\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2X NVENCODE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1X NVENCODE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCamera\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eHSB camera via QSFP slot\u003cbr\u003eUSB camera\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eUp to 20 cameras via HSB\u003cbr\u003eUp to 6 cameras through 16x lanes MIPI CSI-2\u003cbr\u003eUp to 32 cameras using Virtual Channels\u003cbr\u003eC-PHY 2.1 (10.25 Gbps)\u003cbr\u003eD-PHY 2.1 (40 Gbps)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003ePCIe*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eM.2 Key M slot with x4 PCIe Gen5\u003cbr\u003eM.2 Key E slot with x1 PCIe Gen5\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eUp to Gen5 (x8 lanes)\u003cbr\u003eRoot port only—C1 (x1) and C3 (x2)\u003cbr\u003eRoot Point or Endpoint—C2 (x1), C4 (x8), and C5 (x4)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eUSB*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e2x USB-A (3.2 Gen2)\u003cbr\u003e2x USB-C (3.1)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003exHCI host controller with integrated PHY (up to)\u003cbr\u003e3x USB 3.2\u003cbr\u003e4x USB 2.0\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eNetworking*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1x 5GBe RJ45 connector\u003cbr\u003e1x QSFP28 (4x 25 GbE)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e4x 25 GbE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e3x 25 Gbe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eDisplay\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1x HDMI 2.0b\u003cbr\u003e1x DisplayPort 1.4a\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e4x shared HDMI2.1\u003cbr\u003eVESA DisplayPort 1.4a—HBR3, MST\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eOther I\/O\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eQSFP connector\u003cbr\u003eM.2 Key E expansion slot (WLAN\/BT, x1 PCIe, USB2.0, UART, I2C, I2S)\u003cbr\u003eM.2 Key M connector (NVMe for storage)\u003cbr\u003ePCIe x4 lane, I2C, PCIe x2 lane\u003cbr\u003e2x 13-pin CAN header\u003cbr\u003e2x 6-pin automation header\u003cbr\u003eLED\u003cbr\u003eJTAG connector (2x 5-pin header)\u003cbr\u003e1x fan connector —12V, PWM, and Tach\u003cbr\u003eAudio panel header (2x 5-pin)\u003cbr\u003eMicrofit power jack\u003cbr\u003eRTC backup battery connector 2-pin\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e5x I2S\/2x audio hub (AHUB), 2x DMIC, 4x UART, 4x CAN, 3x SPI, 13x I2C, 6x PWM outputs\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e5x I2S\/2x audio hub (AHUB), 2x DMIC, 4x UART, 3x SPI, 13x I2C, 6x PWM outputs**\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003ePower\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–130 W\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–130 W\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–70 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eMechanical\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e243.19 mm x 112.40 mm x 56.88 mm\u003cbr\u003eThermal Transfer Plate (TTP) and optional fan or heat sink\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e100 mm x 87 mm\u003cbr\u003e699-pin B2B connector\u003cbr\u003eIntegrated Thermal Transfer Plate (TTP) with heatpipe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"3\"\u003eDownloads\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-y94g1tg\" data-tab=\"3\"\u003e \u003csection data-id=\"gkncdjw\" class=\"bdf-e bdf-e-gkncdjw bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"jmazoxr\" class=\"bdf-e bdf-e-jmazoxr bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"rnmjjsw\" class=\"bdf-e bdf-e-rnmjjsw bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/cdn.shopify.com\/s\/files\/1\/1913\/0443\/files\/robotics-datasheet-jetson-thor-module-t5000-t4000-nvidia-us-4554251.pdf?v=1772488109\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eDatasheet\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"ff3v58i\" class=\"bdf-e bdf-e-ff3v58i bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/downloads\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eProduct Download\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"wnxmedk\" class=\"bdf-e bdf-e-wnxmedk bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/devtalk.nvidia.com\/default\/board\/139\/embedded-systems\/1\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eProduct FAQ\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"7j4k7vo\" class=\"bdf-e bdf-e-7j4k7vo bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/develop\/tools\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eTools\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"pt6hogd\" class=\"bdf-e bdf-e-pt6hogd bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/learn\/tutorials\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eTutoriais\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e\u003c\/div\u003e","published_at":"2025-07-29T19:36:01+01:00","created_at":"2025-07-29T19:04:28+01:00","vendor":"NVIDIA","type":"Módulos NVIDIA","tags":[],"price":325800,"price_min":325800,"price_max":325800,"available":true,"price_varies":false,"compare_at_price":null,"compare_at_price_min":0,"compare_at_price_max":0,"compare_at_price_varies":false,"variants":[{"id":52327125582164,"title":"Default Title","option1":"Default Title","option2":null,"option3":null,"sku":"900-13834-0080-000","requires_shipping":true,"taxable":true,"featured_image":null,"available":true,"name":"Módulo NVIDIA Jetson T5000","public_title":null,"options":["Default Title"],"price":325800,"weight":1000,"compare_at_price":null,"inventory_quantity":0,"inventory_management":null,"inventory_policy":"deny","barcode":"900-13834-0080-000","requires_selling_plan":false,"selling_plan_allocations":[]}],"images":["\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50003.png?v=1753812491","\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50001.png?v=1753812492","\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50002.png?v=1753812491"],"featured_image":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50003.png?v=1753812491","options":["Title"],"media":[{"alt":"Módulo NVIDIA Jetson T5000 - 900-13834-0080-000","id":59338479403348,"position":1,"preview_image":{"aspect_ratio":1.0,"height":400,"width":400,"src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50003.png?v=1753812491"},"aspect_ratio":1.0,"height":400,"media_type":"image","src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50003.png?v=1753812491","width":400},{"alt":"Módulo NVIDIA Jetson T5000 - 900-13834-0080-000","id":59338479436116,"position":2,"preview_image":{"aspect_ratio":1.0,"height":400,"width":400,"src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50001.png?v=1753812492"},"aspect_ratio":1.0,"height":400,"media_type":"image","src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50001.png?v=1753812492","width":400},{"alt":"Módulo NVIDIA Jetson T5000 - 900-13834-0080-000","id":59338479468884,"position":3,"preview_image":{"aspect_ratio":1.0,"height":400,"width":400,"src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50002.png?v=1753812491"},"aspect_ratio":1.0,"height":400,"media_type":"image","src":"\/\/guedeslopes.pt\/cdn\/shop\/files\/JetsonT50002.png?v=1753812491","width":400}],"requires_selling_plan":false,"selling_plan_groups":[],"content":"\u003cdiv id=\"buildify\" class=\"buildify buildify-12156019278164\"\u003e\u003cdiv class=\"bdf-innr\"\u003e \u003cdiv class=\"bdf-s-wrp\"\u003e \u003csection data-id=\"ywqzfdc\" class=\"bdf-e bdf-e-ywqzfdc bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-top-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"awuaawd\" class=\"bdf-e bdf-e-awuaawd bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"lfuzqeo\" class=\"bdf-e bdf-e-lfuzqeo bdf-w bdf-w-heading\" data-el=\"heading.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003ch2 class=\"bdf-h-t bdf-sz-medium\"\u003eModulo NVIDIA Jetson T5000\u003c\/h2\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"zgedg2q\" class=\"bdf-e bdf-e-zgedg2q bdf-w bdf-w-heading\" data-el=\"heading.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003ch2 class=\"bdf-h-t bdf-sz-small\"\u003e900-13834-0080-000\u003c\/h2\u003e \u003c\/div\u003e \u003c\/div\u003e \u003csection data-id=\"dzrjrze\" class=\"bdf-e bdf-e-dzrjrze bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"jachcmk\" class=\"bdf-e bdf-e-jachcmk bdf-c bdf-c-s-100 bdf-inner-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"ulccafn\" class=\"bdf-e bdf-e-ulccafn bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\u003cul\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eMódulo NVIDIA Jetson T5000\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003e2070 FP4 TFLOPS Computação de IA powered by Blackwell GPU\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eCPU ARM Neoverse de 4 núcleos\u003c\/span\u003e\u003c\/li\u003e\n\u003cli\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eMemória LPDDR5X de 128 GB\u003c\/span\u003e\u003c\/li\u003e\n\u003c\/ul\u003e\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003cdiv data-id=\"vegfcxn\" class=\"bdf-e bdf-e-vegfcxn bdf-tbs-v-horizontal bdf-w bdf-w-tabs\" data-el=\"tabs.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-tbs bdf-rcv-tb-w\" role=\"tablist\"\u003e \u003cdiv class=\"bdf-tbs-wrp\" role=\"tab\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"1\"\u003eVista Rápida\u003c\/div\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"2\"\u003eCaracterísticas\u003c\/div\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-dt-t\" data-tab=\"3\"\u003eDownloads\u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tbs-cnt-wrp bdf-rcv-tb-wrp\" role=\"tabpanel\"\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"1\"\u003eVista Rápida\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-kfp0noq\" data-tab=\"1\"\u003e \u003csection data-id=\"kxfflkj\" class=\"bdf-e bdf-e-kxfflkj bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"qkyonbf\" class=\"bdf-e bdf-e-qkyonbf bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"heduntc\" class=\"bdf-e bdf-e-heduntc bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003e\u003cstrong\u003eA Plataforma Definitiva para AI Física e Robótica\u003c\/strong\u003e\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO NVIDIA® Jetson T5000™ oferece o mais alto nível de desempenho para AI física e robótica. Você obtém até 2070 FP4 TFLOPS de computação em AI e 128 GB de memória com potência configurável entre 40 W e 130 W. Comparado ao AGX Orin™, ele fornece mais de 7.5x mais computação em AI do que o NVIDIA AGX Orin™, com 3.5x melhor eficiência energética.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eEste sistema-em-módulo apresenta uma GPU com arquitetura NVIDIA Blackwell, um motor de transformador e suporte para GPU de Multi-Instância (MIG) para executar facilmente os mais recentes modelos generativos de AI. O NVIDIA Jetson Thor acelera aplicações multi-sensor em tempo real e de baixa latência com uma CPU Arm® Neoverse®-V3AE de 14 núcleos, 4x rede 25 GbE e extensas opções de E\/S para fusão de sensores. Também inclui um conjunto de aceleradores, como um Acelerador de Visão Programável (PVA) de terceira geração, codificadores e decodificadores duplos, um acelerador de fluxo óptico, e mais.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO Jetson Thor pertence a uma nova classe de computadores robóticos, projetados do zero para alimentar aplicações AI físicas de próxima geração. Ele suporta uma ampla gama de modelos generativos de AI, desde modelos de Ação de Linguagem Visual (VLA) como o NVIDIA Isaac™ GR00T para humanos até todos os modelos LLMs e VLMs populares. Para oferecer uma experiência perfeita de nuvem à borda, o Jetson Thor executa a pilha de software AI da NVIDIA para aplicações de AI física, incluindo NVIDIA Isaac para robótica, NVIDIA Metropolis para AI visual agente e NVIDIA Holoscan para processamento de sensores. Você também pode construir agentes de AI na borda usando fluxos de trabalho de AI agente da NVIDIA, como Pesquisa e Resumo de Vídeo (VSS).\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eNosso ecossistema de parceiros oferece todas as placas de transporte, serviços de design, câmaras e outros sensores de que você precisa, bem como software adicional de AI e sistema. Isso permite acelerar o desenvolvimento de soluções em indústrias que vão desde robótica e espaços inteligentes, até retalho, industrial, médica e muito mais.\u003c\/span\u003e\u003c\/p\u003e\n\u003cp\u003e\u003cspan style=\"font-size: 16px; line-height: 26px; display: inline-block;\"\u003eO módulo Jetson Thor oferece desempenho pronto para produção, computação massiva de AI e capacidades de sensores para aplicações de AI física em um formato compacto. Isso torna-o a plataforma ideal para desenvolvedores que procuram desbloquear novas possibilidades para robótica humanoide e outras aplicações de AI física.\u003c\/span\u003e\u003c\/p\u003e\n\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"2\"\u003eCaracterísticas\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-1jpf9wf\" data-tab=\"2\"\u003e \u003csection data-id=\"19sojk8\" class=\"bdf-e bdf-e-19sojk8 bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"7fzd5zq\" class=\"bdf-e bdf-e-7fzd5zq bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"27otq25\" class=\"bdf-e bdf-e-27otq25 bdf-w bdf-w-text-editor\" data-el=\"text-editor.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-txt-edt bdf-clr\"\u003e\n\u003cdiv class=\"nv-title text h--medium aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-title-6f099a7d2d\" class=\"general-container-text \"\u003e\u003cdiv class=\"text-center lap-text-center tab-text-center mob-text-center\"\u003e\u003ch2 class=\"title \"\u003eNVIDIA Jetson Thor Series\u003c\/h2\u003e\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv class=\"nv-separator separator aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-separator-60a56b94ec\" class=\"cmp-separator\"\u003e \u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv class=\"nv-text text aem-GridColumn aem-GridColumn--default--12\"\u003e\u003cdiv id=\"nv-text-1228782b3d\" class=\"general-container-text \"\u003e\u003cdiv class=\"text-left lap-text-left tab-text-left mob-text-left\"\u003e\u003cdiv class=\"description \"\u003e\n\u003cdiv id=\"jetson-prod-module-table-topscroll\"\u003e\u003cdiv class=\"scroll-div1\"\u003e \u003c\/div\u003e\u003c\/div\u003e\n\u003cdiv id=\"jetson-prod-module-table\" class=\"jetson-specs-table scrolling expanded\"\u003e\u003cdiv class=\"scroll-div2\"\u003e\u003ctable border=\"0\" cellspacing=\"0\" cellpadding=\"0\" align=\"center\"\u003e\u003ctbody\u003e\n\u003ctr class=\"head border\"\u003e\n\u003ctd class=\"titlehead tableCLdata1\" width=\"10%\"\u003e \u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson AGX Thor Developer Kit\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson T5000\u003c\/strong\u003e\u003c\/td\u003e\n\u003ctd class=\"titlehead tableCMdata1\" width=\"18%\"\u003e\u003cstrong\u003eJetson T4000\u003c\/strong\u003e\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eAI Performance\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2070 TFLOPS (FP4—Sparse)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1200 TFLOPS (FP4—Sparse)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eGPU\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2560-core NVIDIA Blackwell architecture GPU with fifth-gen Tensor Cores\u003cbr\u003eMulti-Instance GPU (MIG) with 10 TPCs\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1536-core NVIDIA Blackwell architecture GPU with fifth-gen Tensor Cores\u003cbr\u003eMulti-Instance GPU (MIG) with 6 TPCs\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eGPU Max Frequency\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e1.57 GHz\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1.53 GHz\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCPU\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e14-core Arm® Neoverse®-V3AE 64-bit CPU\u003cbr\u003e1 MB L2 cache per core\u003cbr\u003e16 MB shared system L3 cache\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e12-core Arm® Neoverse®-V3AE 64-bit CPU\u003cbr\u003e1 MB L2 cache per core\u003cbr\u003e16 MB shared system L3 cache\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCPU Max Frequency\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"3\"\u003e2.6 GHz\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003e\u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/pva\" target=\"_blank\" rel=\"noopener\"\u003eVision Accelerator\u003c\/a\u003e\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"3\"\u003e1x PVA v3\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eMemory\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e128 GB 256-bit LPDDR5X\u003cbr\u003e273 GB\/s\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e64 GB 256-bit LPDDR5X\u003cbr\u003e273 GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eStorage\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1 TB NVMe M.2 Key M Slot\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eSupports NVMe through PCIe\u003cbr\u003eSupports SSD through USB3.2\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eVideo Encode\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2X NVENCODE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1X NVENCODE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eVideo Decode\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e2X NVENCODE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1X NVENCODE\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eCamera\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eHSB camera via QSFP slot\u003cbr\u003eUSB camera\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eUp to 20 cameras via HSB\u003cbr\u003eUp to 6 cameras through 16x lanes MIPI CSI-2\u003cbr\u003eUp to 32 cameras using Virtual Channels\u003cbr\u003eC-PHY 2.1 (10.25 Gbps)\u003cbr\u003eD-PHY 2.1 (40 Gbps)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003ePCIe*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eM.2 Key M slot with x4 PCIe Gen5\u003cbr\u003eM.2 Key E slot with x1 PCIe Gen5\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003eUp to Gen5 (x8 lanes)\u003cbr\u003eRoot port only—C1 (x1) and C3 (x2)\u003cbr\u003eRoot Point or Endpoint—C2 (x1), C4 (x8), and C5 (x4)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eUSB*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e2x USB-A (3.2 Gen2)\u003cbr\u003e2x USB-C (3.1)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003exHCI host controller with integrated PHY (up to)\u003cbr\u003e3x USB 3.2\u003cbr\u003e4x USB 2.0\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eNetworking*\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1x 5GBe RJ45 connector\u003cbr\u003e1x QSFP28 (4x 25 GbE)\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e4x 25 GbE\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e3x 25 Gbe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eDisplay\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e1x HDMI 2.0b\u003cbr\u003e1x DisplayPort 1.4a\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e4x shared HDMI2.1\u003cbr\u003eVESA DisplayPort 1.4a—HBR3, MST\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eOther I\/O\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003eQSFP connector\u003cbr\u003eM.2 Key E expansion slot (WLAN\/BT, x1 PCIe, USB2.0, UART, I2C, I2S)\u003cbr\u003eM.2 Key M connector (NVMe for storage)\u003cbr\u003ePCIe x4 lane, I2C, PCIe x2 lane\u003cbr\u003e2x 13-pin CAN header\u003cbr\u003e2x 6-pin automation header\u003cbr\u003eLED\u003cbr\u003eJTAG connector (2x 5-pin header)\u003cbr\u003e1x fan connector —12V, PWM, and Tach\u003cbr\u003eAudio panel header (2x 5-pin)\u003cbr\u003eMicrofit power jack\u003cbr\u003eRTC backup battery connector 2-pin\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e5x I2S\/2x audio hub (AHUB), 2x DMIC, 4x UART, 4x CAN, 3x SPI, 13x I2C, 6x PWM outputs\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e5x I2S\/2x audio hub (AHUB), 2x DMIC, 4x UART, 3x SPI, 13x I2C, 6x PWM outputs**\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003ePower\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–130 W\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–130 W\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e40 W–70 W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd class=\"tableCLdata1\"\u003eMechanical\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\"\u003e243.19 mm x 112.40 mm x 56.88 mm\u003cbr\u003eThermal Transfer Plate (TTP) and optional fan or heat sink\u003c\/td\u003e\n\u003ctd class=\"tableCMdata1\" colspan=\"2\"\u003e100 mm x 87 mm\u003cbr\u003e699-pin B2B connector\u003cbr\u003eIntegrated Thermal Transfer Plate (TTP) with heatpipe\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\u003c\/div\u003e\n\u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv class=\"bdf-tb\"\u003e \u003cdiv class=\"bdf-tb-t bdf-tb-ph-t\" data-tab=\"3\"\u003eDownloads\u003c\/div\u003e \u003cdiv class=\"bdf-rcv-tb-cnt bdf-tb-cnt bdf-s-wrp bdf-clr bdf-rptr-itm-y94g1tg\" data-tab=\"3\"\u003e \u003csection data-id=\"gkncdjw\" class=\"bdf-e bdf-e-gkncdjw bdf-s-boxed bdf-s-h-default bdf-s-h-default bdf-s bdf-inner-s\" data-el=\"section\"\u003e \u003cdiv class=\"bdf-ctr bdf-c-gap-default\"\u003e \u003cdiv class=\"bdf-r\"\u003e \u003cdiv data-id=\"jmazoxr\" class=\"bdf-e bdf-e-jmazoxr bdf-c bdf-c-s-100 bdf-top-c\" data-el=\"column\"\u003e \u003cdiv class=\"bdf-c-wrp bdf-e-ppl\"\u003e \u003cdiv class=\"bdf-w-wrp\"\u003e \u003cdiv data-id=\"rnmjjsw\" class=\"bdf-e bdf-e-rnmjjsw bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/cdn.shopify.com\/s\/files\/1\/1913\/0443\/files\/robotics-datasheet-jetson-thor-module-t5000-t4000-nvidia-us-4554251.pdf?v=1772488109\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eDatasheet\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"ff3v58i\" class=\"bdf-e bdf-e-ff3v58i bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/downloads\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eProduct Download\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"wnxmedk\" class=\"bdf-e bdf-e-wnxmedk bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/devtalk.nvidia.com\/default\/board\/139\/embedded-systems\/1\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eProduct FAQ\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"7j4k7vo\" class=\"bdf-e bdf-e-7j4k7vo bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/develop\/tools\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eTools\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003cdiv data-id=\"pt6hogd\" class=\"bdf-e bdf-e-pt6hogd bdf-w bdf-w-button\" data-el=\"button.default\"\u003e \u003cdiv class=\"bdf-w-ctr\"\u003e \u003cdiv class=\"bdf-btn-wrp\"\u003e \u003ca href=\"https:\/\/developer.nvidia.com\/embedded\/learn\/tutorials\" class=\"bdf-btn-link bdf-btn bdf-sz-md\" target=\"_blank\"\u003e \u003cspan class=\"bdf-btn-cnt-wrp\"\u003e \u003cspan class=\"bdf-aln-icon-right bdf-btn-icon\"\u003e \u003ci class=\"fa fa-angle-right\"\u003e\u003c\/i\u003e \u003c\/span\u003e \u003cspan class=\"bdf-btn-text\"\u003eTutoriais\u003c\/span\u003e \u003c\/span\u003e \u003c\/a\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/div\u003e \u003c\/section\u003e \u003c\/div\u003e \u003c\/div\u003e\u003c\/div\u003e"}