{"product_id":"nvidia-jetson-agx-thor-developer-kit","title":"NVIDIA Jetson AGX Thor 開發套件 128GB LPDDR5x，2070 FP4 TFLOPS，1TB NVMe SSD，Wi‑Fi 6E","description":"\u003ch2\u003e概述\u003c\/h2\u003e\u003cp\u003eNVIDIA Jetson AGX Thor \u003cstrong\u003e開發套件\u003c\/strong\u003e 是一個基於Blackwell架構的計算平台，用於機器人和邊緣AI開發。它提供高達\u003cstrong\u003e2070 FP4 TFLOPS\u003c\/strong\u003e 的AI計算能力，並被指定為比上一代Jetson AGX Orin高出\u003cstrong\u003e7.5倍\u003c\/strong\u003e 的性能，具有\u003cstrong\u003e3.5倍\u003c\/strong\u003e 的能效提升。該套件配備\u003cstrong\u003e128GB LPDDR5x\u003c\/strong\u003e 記憶體（256位元），具有\u003cstrong\u003e273GB\/s\u003c\/strong\u003e 的頻寬，以支持大型Transformer推理、高並發視頻編碼\/解碼和多傳感器數據融合。\u003c\/p\u003e\u003cp\u003eBlackwell \u003cstrong\u003e多實例GPU (MIG)\u003c\/strong\u003e 得到支持，允許單個GPU被分割成多個獨立實例，以便同時處理環境感知、語言交互和行動規劃等工作負載。該平台與CUDA生態系統集成，並支持NVIDIA軟體堆疊，包括\u003cstrong\u003eIsaac\u003c\/strong\u003e, \u003cstrong\u003eIsaac GR00T\u003c\/strong\u003e, \u003cstrong\u003eMetropolis\u003c\/strong\u003e, 和\u003cstrong\u003eHoloscan\u003c\/strong\u003e. \u003c\/p\u003e\u003ch2\u003e主要特點\u003c\/h2\u003e\u003cul\u003e\n\u003cli\u003e\n\u003cstrong\u003eAI性能：\u003c\/strong\u003e 高達2070 FP4 TFLOPS\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eGPU架構：\u003c\/strong\u003e NVIDIA Blackwell\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003eCPU：\u003c\/strong\u003e 14核Arm Neoverse V3AE\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e記憶體：\u003c\/strong\u003e 128GB LPDDR5x，256位元；記憶體頻寬273GB\/s\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e功耗：\u003c\/strong\u003e 40–130W（2070 FP4 TFLOPS操作時也標示為130W）\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e儲存（註明）：\u003c\/strong\u003e 1TB NVMe SSD；提供SSD外殼以寫入系統映像\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e無線（板載）：\u003c\/strong\u003e Wi‑Fi（2.4GHz\/5GHz\/6GHz）高達2402Mbps；藍牙5。3; 在出貨前安裝在設備內\u003c\/li\u003e\n\u003cli\u003e\n\u003cstrong\u003e系統映像（註記）：\u003c\/strong\u003e SSD 隨附預裝系統映像；系統映像提到 AI 視覺功能、ROS 案例、離線大型模型案例和 OpenClaw 課程內容\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch2\u003e規格\u003c\/h2\u003e\u003ctable\u003e\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd\u003e產品\u003c\/td\u003e\n\u003ctd\u003eJetson AGX Thor 開發者套件\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eGPU\u003c\/td\u003e\n\u003ctd\u003eNVIDIA Jetson T5000 with Blackwell GPU\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eAI 計算\u003c\/td\u003e\n\u003ctd\u003e高達 2070 FP4 TFLOPS\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eCPU\u003c\/td\u003e\n\u003ctd\u003e14 核心 (Arm Neoverse V3AE)\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e記憶體\u003c\/td\u003e\n\u003ctd\u003e128GB LPDDR5x, 256-bit\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e記憶體頻寬\u003c\/td\u003e\n\u003ctd\u003e273GB\/s\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e功耗\u003c\/td\u003e\n\u003ctd\u003e40–130W\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e散熱\u003c\/td\u003e\n\u003ctd\u003e散熱片和風扇（風扇和散熱器）\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e存儲（註記）\u003c\/td\u003e\u003ctd\u003e1TB NVMe SSD\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e無線\u003c\/td\u003e\n\u003ctd\u003eWi‑Fi 2。4GHz\/5GHz\/6GHz 高達 2402Mbps；BT 5.3；Wi‑Fi 6E 模組已註明\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e外部 I\/O（已註明）\u003c\/td\u003e\n\u003ctd\u003e2x USB‑A；2x USB‑C；RJ45 連接器（5GbE）；DP 連接器；HDMI 2.1 連接器；QSFP28 連接器\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e按鈕\/指示燈\u003c\/td\u003e\n\u003ctd\u003e電源按鈕；閃爍按鈕；重置按鈕；電源指示燈\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e內部接頭\/介面（已註明）\u003c\/td\u003e\n\u003ctd\u003e音頻（10Pin）；自動化介面（12Pin）；CAN Bus（2x13Pin）；USB Type‑C 調試介面；JTAG（10Pin）；RTC（2Pin）\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e機箱尺寸\u003c\/td\u003e\n\u003ctd\u003e243 x 112.40 x 56.88 mm（單位：mm）；另註明長度 240。86 mm\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003e3D模型\u003c\/td\u003e\n\u003ctd\u003e將提供3D模型文件\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\u003c\/table\u003e\u003ch2\u003e性能數據（如所述）\u003c\/h2\u003e\u003ch3\u003e生成推理比較說明\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e與Jetson AGX Orin相比，Jetson AGX Thor在生成推理方面提供高達\u003cstrong\u003e5倍\u003c\/strong\u003e的速度提升。 \u003c\/li\u003e\n\u003cli\u003e使用\u003cstrong\u003eFP4\u003c\/strong\u003e和 \u003cstrong\u003e推測解碼\u003c\/strong\u003e, 據稱開發者在Jetson AGX Thor上可實現額外\u003cstrong\u003e2倍\u003c\/strong\u003e的性能加速。 \u003c\/li\u003e\n\u003cli\u003e實時多模態AI示例說明：使用Qwen2.5‑VL‑3B VLM和Llama3.2 3B LLM，初始Token (TTFT)響應據稱低於\u003cstrong\u003e200毫秒\u003c\/strong\u003e, 且每個Token (TPOT)輸出時間據稱遠低於\u003cstrong\u003e50毫秒\u003c\/strong\u003e. \u003c\/li\u003e\u003c\/ul\u003e\u003ch3\u003e令牌\/秒 表格 (Jetson AGX Orin vs Jetson AGX Thor)\u003c\/h3\u003e\u003ctable\u003e\n\u003cthead\u003e\u003ctr\u003e\n\u003cth\u003e系列\u003c\/th\u003e\n\u003cth\u003e型號\u003c\/th\u003e\n\u003cth\u003eJetson AGX Orin (令牌\/秒)\u003c\/th\u003e\n\u003cth\u003eJetson AGX Thor (令牌\/秒)\u003c\/th\u003e\n\u003cth\u003e加速倍數\u003c\/th\u003e\n\u003c\/tr\u003e\u003c\/thead\u003e\n\u003ctbody\u003e\n\u003ctr\u003e\n\u003ctd\u003eLlama\u003c\/td\u003e\n\u003ctd\u003eLlama 3.1 8B\u003c\/td\u003e\n\u003ctd\u003e112.33\u003c\/td\u003e\n\u003ctd\u003e150.8\u003c\/td\u003e\n\u003ctd\u003e1.34\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eLlama\u003c\/td\u003e\n\u003ctd\u003eLlama 3.3 70B\u003c\/td\u003e\n\u003ctd\u003e7.38\u003c\/td\u003e\n\u003ctd\u003e12.64\u003c\/td\u003e\n\u003ctd\u003e1.71\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eQwen\u003c\/td\u003e\n\u003ctd\u003eQwen3-30B-A3B\u003c\/td\u003e\n\u003ctd\u003e76.69\u003c\/td\u003e\n\u003ctd\u003e226.42\u003c\/td\u003e\n\u003ctd\u003e2.95\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eQwen\u003c\/td\u003e\n\u003ctd\u003eQwen3-32B\u003c\/td\u003e\n\u003ctd\u003e16.84\u003c\/td\u003e\n\u003ctd\u003e79.1\u003c\/td\u003e\n\u003ctd\u003e4.7\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eDeepSeek\u003c\/td\u003e\n\u003ctd\u003eDeepSeek-R1-Distill-Qwen-7B\u003c\/td\u003e\n\u003ctd\u003e180.41\u003c\/td\u003e\n\u003ctd\u003e304.76\u003c\/td\u003e\n\u003ctd\u003e1.69\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eDeepSeek\u003c\/td\u003e\n\u003ctd\u003eDeepSeek-R1-Distill-Qwen-32B\u003c\/td\u003e\n\u003ctd\u003e16.96\u003c\/td\u003e\n\u003ctd\u003e82.63\u003c\/td\u003e\n\u003ctd\u003e4.87\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eQwen (VLM)\u003c\/td\u003e\n\u003ctd\u003eQwen2.5-VL-3B\u003c\/td\u003e\n\u003ctd\u003e216\u003c\/td\u003e\n\u003ctd\u003e356.86\u003c\/td\u003e\n\u003ctd\u003e1.65\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eQwen (VLM)\u003c\/td\u003e\n\u003ctd\u003eQwen2.5-VL-7B\u003c\/td\u003e\n\u003ctd\u003e154.02\u003c\/td\u003e\n\u003ctd\u003e252\u003c\/td\u003e\n\u003ctd\u003e1.64\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eLlama (VLM)\u003c\/td\u003e\n\u003ctd\u003eLlama 3.2 11B Vision\u003c\/td\u003e\n\u003ctd\u003e44.22\u003c\/td\u003e\n\u003ctd\u003e69.63\u003c\/td\u003e\n\u003ctd\u003e1.57\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eGR00T (VLA)\u003c\/td\u003e\n\u003ctd\u003eGR00T N1\u003c\/td\u003e\n\u003ctd\u003e18.5\u003c\/td\u003e\n\u003ctd\u003e46.7\u003c\/td\u003e\n\u003ctd\u003e2.52\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003ctr\u003e\n\u003ctd\u003eGR00T (VLA)\u003c\/td\u003e\n\u003ctd\u003eGR00T N1.5\u003c\/td\u003e\n\u003ctd\u003e15.2\u003c\/td\u003e\n\u003ctd\u003e41.5\u003c\/td\u003e\n\u003ctd\u003e2.74\u003c\/td\u003e\n\u003c\/tr\u003e\n\u003c\/tbody\u003e\n\u003c\/table\u003e\u003ch2\u003e端口 \u0026amp; 功能分佈（如標示）\u003c\/h2\u003e \u003cul\u003e\n\u003cli\u003e直流電源供應\u003c\/li\u003e\n\u003cli\u003eUSB Type‑C 介面（可用作電源輸入介面）\u003c\/li\u003e\n\u003cli\u003eHDMI\u003c\/li\u003e\n\u003cli\u003eUSB Type‑A 介面\u003c\/li\u003e\n\u003cli\u003eDP 介面\u003c\/li\u003e\n\u003cli\u003e乙太網埠\u003c\/li\u003e\n\u003cli\u003eQSFP28 連接器\u003c\/li\u003e\n\u003cli\u003e風扇和散熱器\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch2\u003e系統映像 \u0026amp; 軟體環境（如所述）\u003c\/h2\u003e \u003cul\u003e\n\u003cli\u003e\u003cstrong\u003eSSD 已預裝系統映像。\u003c\/strong\u003e\u003c\/li\u003e\n\u003cli\u003eJetson AGX Thor 在出貨前附有 \u003cstrong\u003e1TB 空 SSD\u003c\/strong\u003e，並提供 \u003cstrong\u003e SSD 外殼\u003c\/strong\u003e 用於寫入系統映像檔案。\u003c\/li\u003e \n\u003cli\u003e也提到了一個 \u003cstrong\u003e2TB SSD\u003c\/strong\u003e，其中包含系統映像檔案。如果選擇2TB SSD版本套件，官方套件中附帶的原始1TB SSD將被保留；包裝中的2TB SSD必須由用戶自行更換，且默認不會移除原始包裝。\u003c\/li\u003e \n\u003cli\u003e\n\u003cstrong\u003e基於Ubuntu 24.04的\u003c\/strong\u003e Jetson系統映像，支持\u003cstrong\u003e NVIDIA CUDA 13.0\u003c\/strong\u003e, \u003cstrong\u003eTensorRT 10.13\u003c\/strong\u003e, \u003cstrong\u003ecuDNN 9.12\u003c\/strong\u003e, 和\u003cstrong\u003eOpenCV 4.13\u003c\/strong\u003e. \u003c\/li\u003e\n\u003c\/ul\u003e\u003ch2\u003e可選 \/ 支持的模塊（如描述）\u003c\/h2\u003e\u003ch3\u003eAI大型模型語音模塊（支持聲明）\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e高性能麥克風和腔體揚聲器\u003c\/li\u003e\n\u003cli\u003e遠場拾音、回聲消除和環境噪聲抑制\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch3\u003eUSB工業相機（支持聲明）\u003c\/h3\u003e\u003cul\u003e\n\u003cli\u003e5MP\u003c\/li\u003e\n\u003cli\u003e支持在2592 x 1944下穩定輸出30fps\u003c\/li\u003e\n\u003cli\u003e105°對角視場（水平81。8°，垂直66°C（如所述）\u003c\/li\u003e\n\u003cli\u003e無失真影像；防塵外殼（如所述）\u003c\/li\u003e\n\u003c\/ul\u003e\u003ch2\u003e教學影片\u003c\/h2\u003e\u003cp\u003e\u003ciframe height=\"751\" src=\"https:\/\/www.youtube.com\/embed\/-20a1fotbZA\" title=\"Jetson AGX Thor: 2070 TFLOPS of Edge AI Power (Full Demo)\" width=\"1335\"\u003e\u003c\/iframe\u003e\u003c\/p\u003e\u003ch2\u003e支援\u003c\/h2\u003e\u003cp\u003e如需技術問題或選擇Jetson AGX Thor開發套件的正確配置的幫助，請聯繫支援\u003ca href=\"https:\/\/rcdrone.top\/\"\u003ehttps:\/\/rcdrone.top\/\u003c\/a\u003e或\u003ca href=\"mailto:support@rcdrone.top\"\u003e support@rcdrone.top\u003c\/a\u003e. \u003c\/p\u003e\u003ch2 id=\"details\"\u003e詳情\u003c\/h2\u003e\u003cdiv class=\"details-gallery\"\u003e\n\u003cimg alt=\"NVIDIA Jetson AGX Thor developer kit with 1TB NVMe system image note and onboard Wi‑Fi 6E card detail\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4.jpg?v=1782281427\"\u003e\u003cp\u003e隨附的1TB NVMe SSD附帶預裝的系統映像，並且板載無線卡支持Wi‑Fi 6E和藍牙5.3。\u003c\/p\u003e\n\u003cimg alt=\"Jetson AGX Thor course and tutorial portal screen, listing developer kit learning categories and resources\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_cf2878a0-9f1b-4390-98f4-5b4c97cba6c9.jpg?v=1782281433\"\u003e\u003cimg alt=\"Jetson AGX Thor training outline with ROS, vision, LLM, and OpenClaw course modules listed in a long table\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_827b4494-4e1a-476c-8b46-ff97c7257ae5.jpg?v=1782281439\"\u003e\u003cimg alt=\"Folders for detailed tutorials and open-source code, plus basic setup, vision, and ROS advanced tutorial lists\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_23740e96-5bcb-49cd-b6d8-1e44ed4eb5a0.jpg?v=1782281446\"\u003e\u003cimg alt=\"Offline and online AI large model development materials with demos for text-to-picture, vision, and localization tasks\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_1bf3064d-c203-4573-99b9-5080d7c42d61.jpg?v=1782281452\"\u003e\u003cimg alt=\"Offline generative AI application materials with example projects like storytelling, cooking assistant, homework, and animation\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_79e3a071-2214-4073-b2d6-e45d706a3fc0.jpg?v=1782281458\"\u003e\u003cimg alt=\"OpenClaw advanced AI development materials with examples for file management, camera control, browser control, and coding\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_8cc625d4-7292-446e-a945-86275f35f03f.jpg?v=1782281465\"\u003e\u003cimg alt=\"Jetson AGX Thor kit connected to AI voice module and USB industrial camera, highlighting supported accessories\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_ee3e36a0-fc30-467e-9211-039c8c78a480.jpg?v=1782281471\"\u003e\u003cp\u003e使用支持的外圍設備（如AI語音模塊和USB工業相機）構建語音助手或視覺管道。\u003c\/p\u003e\n\u003cimg alt=\"Jetson AGX Thor developer kit angled view with rear I\/O ports visible and AI model ecosystem logos above\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_ff8c71db-a82a-42cd-867c-d8db436f32cd.jpg?v=1782281476\"\u003e\u003cimg alt=\"Performance charts comparing Jetson AGX Thor vs Orin for generative reasoning and real-time multimodal token response\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_8367d40c-19ef-4fb0-90fa-1a3cd1800c15.jpg?v=1782281483\"\u003e\u003cp\u003e性能圖表總結了生成推理增益和邊緣推斷工作負載的多模態吞吐量目標。\u003c\/p\u003e\u003cimg alt=\"Jetson AGX Thor benchmark table and key specs callout: up to 2070 FP4 TFLOPS, 128GB LPDDR5x, 14-core CPU\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_e9874dc8-80b9-4291-b15c-6b084274332f.jpg?v=1782281489\"\u003e\u003cp\u003e核心平台亮點包括高達2070 FP4 TFLOPS的AI計算能力、128GB LPDDR5x記憶體和14核Arm CPU。\u003c\/p\u003e \u003cimg alt=\"Robotics and edge AI stack overview for Jetson AGX Thor, with sensor processing, low-latency pipeline, and software icons\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_cf37e8be-768c-4f5f-9aea-8760a6d88b4c.jpg?v=1782281495\"\u003e\u003cimg alt=\"Jetson AGX Thor functional distribution diagram labeling ports, buttons, cooling, and internal headers like CAN and JTAG\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_e41af2f0-4be5-494a-b1ba-7178e1dcef64.jpg?v=1782281501\"\u003e\u003cp\u003e清晰的端口和接頭圖有助於規劃整合，從USB和顯示輸出到CAN總線、調試和其他內部介面。\u003c\/p\u003e \u003cimg alt=\"Jetson AGX Thor dimensions drawing in millimeters with 3D model availability note beneath the chassis rendering\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_0e9d0655-207f-4002-9208-904f5d3cfc0a.jpg?v=1782281507\"\u003e\u003cp\u003e機械圖紙以毫米為單位列出機箱尺寸，並提供用於外殼設計的3D模型文件。\u003c\/p\u003e \u003cimg alt=\"System image screen noting Ubuntu 24.04-based Jetson system with CUDA, TensorRT, cuDNN, and OpenCV versions listed\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_3d0c8588-4b33-4531-8b9e-1d3e91a1368f.jpg?v=1782281514\"\u003e\u003cp\u003e更新的系統映像基於Ubuntu 24.04，並參考CUDA軟體堆疊中的常見加速庫。\u003c\/p\u003e \u003cimg alt=\"Jetson full range comparison table highlighting Jetson AGX Thor column against Orin Nano, Orin NX, AGX Orin, and Xavier\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_109eba99-f3aa-4ced-ac4c-562999f256ff.jpg?v=1782281520\"\u003e\u003cimg alt=\"Continuation of Jetson comparison table with I\/O, networking, display, power, and dimensions across multiple Jetson models\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_2b32d987-a2f7-4961-8eee-93bb92945f25.jpg?v=1782281526\"\u003e\u003cimg alt=\"Jetson AGX Thor developer kit product photos with included power adapter and cables, plus retail box packaging\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_6521bc86-a135-4fba-bc9e-9930ab09f048.jpg?v=1782281531\"\u003e\u003cp\u003e盒內包含：Jetson AGX Thor開發套件、電源供應器和初始啟動所需的電纜。\u003c\/p\u003e\u003cimg alt=\"Shipping list graphic for NVIDIA Jetson AGX Thor 128GB developer kit with power adapter, USB-C cable, optional SSD and accessories\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_1469dd01-c4e7-4e5e-bb60-8fd9e514ad56.jpg?v=1782281538\"\u003e\u003cp\u003eJetson AGX Thor 128GB 開發者套件隨附電源適配器和 USB‑C 數據線，並可選配 1TB\/2TB SSD 和 SSD 外殼，而 Vision Advanced Kit 則增加了 DP‑to‑HDMI 線纜、USB 工業相機等項目。\u003c\/p\u003e\n\u003cimg alt=\"Jetson AGX Thor 128GB developer kit bundle contents with power adapter, USB-C cable, camera, hub, and accessories\" loading=\"lazy\" src=\"https:\/\/rcdrone.top\/cdn\/shop\/files\/Jetson-AGX-Thor-Developer-Kit-with-128GB-RAM2070-FP4_da022928-5444-4766-a00b-a247ab31517e.jpg?v=1782281544\"\u003e\u003cp\u003eJetson AGX Thor 128GB 套件選項將主板與電源適配器、USB‑C 線纜、DP‑to‑HDMI 線、USB 工業相機、SSD 外殼、USB 3.0 集線器以及可選的 NVMe SSD 和觸控屏附加組件捆綁在一起。\u003c\/p\u003e\n\u003c\/div\u003e","brand":"Yahboom","offers":[{"title":"1TB \/ 開發者套件","offer_id":53135903555808,"sku":"6000101391-1","price":6003.8,"currency_code":"USD","in_stock":true},{"title":"1TB \/ Vision 高級套件","offer_id":53135903588576,"sku":"6000101391-2","price":6134.6,"currency_code":"USD","in_stock":true},{"title":"1TB \/ AI大型模型套件","offer_id":53135903621344,"sku":"6000101391-3","price":6170.6,"currency_code":"USD","in_stock":true},{"title":"1TB \/ 終極套裝","offer_id":53135903654112,"sku":"6000101391-4","price":6500.6,"currency_code":"USD","in_stock":true},{"title":"2TB \/ 開發者套件","offer_id":53135903686880,"sku":"6000101391-5","price":6459.8,"currency_code":"USD","in_stock":true},{"title":"2TB \/ Vision 高級套件","offer_id":53135903719648,"sku":"6000101391-6","price":6590.6,"currency_code":"USD","in_stock":true},{"title":"2TB \/ AI大型模型套件","offer_id":53135903752416,"sku":"6000101391-7","price":6626.6,"currency_code":"USD","in_stock":true},{"title":"2TB \/ 終極套裝","offer_id":53135903785184,"sku":"6000101391-8","price":6956.6,"currency_code":"USD","in_stock":true}],"thumbnail_url":"\/\/cdn.shopify.com\/s\/files\/1\/0643\/3210\/3904\/files\/JetsonAGXThorDeveloperKit_4.jpg?v=1782125531","url":"https:\/\/rcdrone.top\/zh\/products\/nvidia-jetson-agx-thor-developer-kit","provider":"RCDrone","version":"1.0","type":"link"}