{"id":62016,"date":"2024-03-19T15:14:00","date_gmt":"2024-03-19T07:14:00","guid":{"rendered":"https:\/\/www.ioiotimes.com\/?p=62016"},"modified":"2024-03-21T00:14:30","modified_gmt":"2024-03-20T16:14:30","slug":"nvidia%e6%8e%a8%e5%87%ba%e7%94%9f%e6%88%90%e5%bc%8f%e4%ba%ba%e5%b7%a5%e6%99%ba%e6%85%a7%e5%be%ae%e6%9c%8d%e5%8b%99%ef%bc%8c%e4%be%9b%e9%96%8b%e7%99%bc%e4%ba%ba%e5%93%a1%e5%9c%a8%e8%a3%9d%e6%9c%89nvidi","status":"publish","type":"post","link":"https:\/\/www.ioiotimes.com\/?p=62016","title":{"rendered":"NVIDIA\u63a8\u51fa\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u5fae\u670d\u52d9\uff0c\u4f9b\u958b\u767c\u4eba\u54e1\u5728\u88dd\u6709NVIDIA CUDA GPU\u7684\u6a5f\u5668\u4e0a\u5efa\u7acb\u53ca\u90e8\u7f72\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u8f14\u52a9\u5925\u4f34"},"content":{"rendered":"\n<ul class=\"wp-block-list\">\n<li><strong>GPU \u52a0\u901f\u7684 NVIDIA NIM \u5fae\u670d\u52d9\u53ca\u96f2\u7aef\u7aef\u9ede\u7684\u5168\u65b0\u76ee\u9304\uff0c\u9069\u7528\u65bc\u9810\u5148\u8a13\u7df4\u7684\u4eba\u5de5\u667a\u6167\u6a21\u578b\uff0c\u5728\u7d93\u904e\u6700\u4f73\u5316\u8abf\u6574\u5f8c\u4fbf\u53ef\u5728\u96f2\u7aef\u3001\u8cc7\u6599\u4e2d\u5fc3\u3001\u5de5\u4f5c\u7ad9\u53ca PC \u4e0a\u6578\u5104\u500b\u652f\u63f4 CUDA \u7684 GPU \u4e0a\u904b\u884c<\/strong><\/li>\n\n\n\n<li><strong>\u4f01\u696d\u53ef\u4f7f\u7528\u5fae\u670d\u52d9\u52a0\u901f\u8cc7\u6599\u8655\u7406\u3001\u5ba2\u88fd\u5316\u5927\u578b\u8a9e\u8a00\u6a21\u578b\u3001\u63a8\u8ad6\u3001\u6aa2\u7d22-\u589e\u5f37\u751f\u6210\u548c Guardrails<\/strong><\/li>\n\n\n\n<li><strong>\u5305\u62ec\u4e3b\u8981\u61c9\u7528\u7a0b\u5f0f\u5e73\u53f0\u63d0\u4f9b\u5546 Cadence\u3001CrowdStrike\u3001SAP\u3001ServiceNow \u7b49\u5ee3\u6cdb\u7684\u4eba\u5de5\u667a\u6167\u751f\u614b\u7cfb\u7d71\u5747\u6709\u63a1\u7528<\/strong><\/li>\n<\/ul>\n\n\n\n<p>NVIDIA \u4eca\u65e5\u63a8\u51fa\u6578\u5341\u7a2e\u4f01\u696d\u7d1a\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\uff08AI\uff09\u5fae\u670d\u52d9\uff0c\u4f01\u696d\u53ef\u4ee5\u4f7f\u7528\u9019\u4e9b\u670d\u52d9\u5728\u81ea\u5df1\u7684\u5e73\u53f0\u4e0a\u5efa\u7acb\u53ca\u90e8\u7f72\u5ba2\u88fd\u5316\u7684\u61c9\u7528\u7a0b\u5f0f\uff0c\u540c\u6642\u53c8\u4fdd\u7559\u5b8c\u5168\u6301\u6709\u548c\u63a7\u5236\u81ea\u6709\u667a\u6167\u8ca1\u7522\u6b0a\u7684\u6b0a\u5229\u3002<\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"559\" src=\"https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia01.png\" alt=\"\" class=\"wp-image-62332\" title=\"\" srcset=\"https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia01.png 1024w, https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia01-300x164.png 300w, https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia01-768x419.png 768w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">\u25b2NVIDIA NIM \u5fae\u670d\u52d9<\/figcaption><\/figure>\n\n\n\n<p><\/p>\n\n\n\n<p>\u5efa\u7acb\u5728 <a href=\"https:\/\/developer.nvidia.com\/about-cuda\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA CUDA<\/a>\u00ae \u00a0\u5e73\u53f0\u4e0a\u7684\u96f2\u7aef\u539f\u751f\u5fae\u670d\u52d9\u76ee\u9304\uff0c\u5176\u4e2d\u5305\u62ec <a href=\"https:\/\/nvidianews.nvidia.com\/news\/generative-ai-microservices-for-developers\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA NIM<sup>TM<\/sup><\/a> \u5fae\u670d\u52d9\uff0c\u7528\u65bc\u5c0d NVIDIA \u53ca\u5176\u5408\u4f5c\u5925\u4f34\u751f\u614b\u7cfb\u7d71\u4e2d\u7684\u4e8c\u5341\u591a\u7a2e\u71b1\u9580AI\u6a21\u578b\u63d0\u4f9b\u6700\u4f73\u7684\u63a8\u8ad6\u7d50\u679c\u3002NVIDIA \u7684\u52a0\u901f\u8edf\u9ad4\u958b\u767c\u5957\u4ef6\u3001\u51fd\u5f0f\u5eab\u53ca\u5de5\u5177\u73fe\u5728\u9084\u80fd\u4f5c\u70ba <a href=\"https:\/\/www.nvidia.com\/en-us\/technologies\/cuda-x\/&#039;\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA CUDA-X<\/a> \u5fae\u670d\u52d9\u958b\u653e\u4f7f\u7528\uff0c\u7528\u65bc\u6aa2\u7d22\u589e\u5f37\u751f\u6210\uff08RAG\uff09\u3001guardrails\u3001\u8cc7\u6599\u8655\u7406\u3001\u9ad8\u6548\u80fd\u904b\u7b97\uff08HPC\uff09\u7b49\u4f5c\u696d\u3002NVIDIA \u4e5f\u5206\u5225\u767c\u8868\u4e86\u4e8c\u5341\u591a\u6b3e\u7528\u65bc<a href=\"https:\/\/nvidianews.nvidia.com\/news\/healthcare-generative-ai-microservices\" target=\"_blank\" rel=\"noreferrer noopener\">\u91ab\u7642\u4fdd\u5065\u9818\u57df\u7684 NIM \u548c CUDA-X \u5fae\u670d\u52d9\u9805\u76ee<\/a>\u3002<\/p>\n\n\n\n<p>\u7d93\u904e\u7cbe\u5fc3\u6311\u9078\u7684\u5fae\u670d\u52d9\u53c8\u70ba NVIDIA \u5168\u7aef\u904b\u7b97\u5e73\u53f0\u52a0\u5165\u4e86\u4e00\u500b\u65b0\u7684\u5c64\u9762\u3002\u9019\u4e00\u5c64\u4ee5\u6a19\u6e96\u5316\u7684\u8def\u5f91\u9023\u63a5\u4e86\u7531\u6a21\u578b\u958b\u767c\u4eba\u54e1\u3001\u5e73\u53f0\u63d0\u4f9b\u5546\u548c\u4f01\u696d\u7d44\u6210\u7684 AI \u751f\u614b\u7cfb\u7d71\uff0c\u4ee5\u904b\u884c\u91dd\u5c0d\u5728\u96f2\u7aef\u3001\u8cc7\u6599\u4e2d\u5fc3\u3001\u5de5\u4f5c\u7ad9\u548c PC \u4e2d\u5b89\u88dd\u7684\u4e0a\u5104\u500b NVIDIA CUDA GPU \u9032\u884c\u6700\u4f73\u5316\u8abf\u6574\u7684\u5ba2\u88fd\u5316 AI \u6a21\u578b\u3002<\/p>\n\n\n\n<p>Adobe\u3001Cadence\u3001CrowdStrike\u3001Getty Images\u3001SAP\u3001ServiceNow \u548c Shutterstock \u662f\u9996\u6279\u4f7f\u7528 <a href=\"https:\/\/blogs.nvidia.com\/blog\/microservices-ai-enterprise\/\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA AI Enterprise 5.0<\/a> \u4e2d\u63d0\u4f9b\u4e4b\u5168\u65b0 NVIDIA \u751f\u6210\u5f0f AI \u5fae\u670d\u52d9\u7684\u4e3b\u8981\u61c9\u7528\u7a0b\u5f0f\u3001\u8cc7\u6599\u548c\u7db2\u8def\u5b89\u5168\u5e73\u53f0\u63d0\u4f9b\u5546\u3002<\/p>\n\n\n\n<p>NVIDIA \u5275\u8fa6\u4eba\u66a8\u57f7\u884c\u9577\u9ec3\u4ec1\u52f3\u8868\u793a\uff1a\u300c\u6210\u719f\u7684\u4f01\u696d\u5e73\u53f0\u5750\u64c1\u8cc7\u6599\u91d1\u7926\uff0c\u9019\u4e9b\u8cc7\u6599\u53ef\u4ee5\u8b8a\u6210\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u7684\u8f14\u52a9\u5925\u4f34\uff08copilot\uff09\u3002\u9019\u4e9b\u8207\u6211\u5011\u7684\u5408\u4f5c\u5925\u4f34\u751f\u614b\u7cfb\u7d71\u5171\u540c\u6253\u9020\u7684\u5bb9\u5668\u5316\u4eba\u5de5\u667a\u6167\u5fae\u670d\u52d9\uff0c\u662f\u5404\u884c\u5404\u696d\u7684\u4f01\u696d\u6210\u70ba\u4eba\u5de5\u667a\u6167\u516c\u53f8\u7684\u57fa\u77f3\u3002\u300d<\/p>\n\n\n\n<p><\/p>\n\n\n\n<figure class=\"wp-block-image size-full\"><img loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"576\" src=\"https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia02.png\" alt=\"\" class=\"wp-image-62333\" title=\"\" srcset=\"https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia02.png 1024w, https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia02-300x169.png 300w, https:\/\/www.ioiotimes.com\/wordpress\/wp-content\/uploads\/2024\/03\/20240320-nvidia02-768x432.png 768w\" sizes=\"(max-width: 1024px) 100vw, 1024px\" \/><figcaption class=\"wp-element-caption\">\u25b2NVIDIA AI Enterprise 5.0<\/figcaption><\/figure>\n\n\n\n<p><\/p>\n\n\n\n<p><strong>NIM\u63a8\u8ad6\u5fae\u670d\u52d9\u5c07\u90e8\u7f72\u6642\u9593\u5f9e\u6578\u9031\u7e2e\u77ed\u81f3\u53ea\u8981\u6578\u5206\u9418\u5373\u53ef\u5b8c\u6210<\/strong><\/p>\n\n\n\n<p>NIM \u5fae\u670d\u52d9\u63d0\u4f9b\u7531 NVIDIA \u63a8\u8ad6\u8edf\u9ad4\uff08\u5305\u62ec Triton Inference Server<sup>TM<\/sup> \u548c TensorRT<sup>TM<\/sup>-LLM\uff09\u9a45\u52d5\u7684\u9810\u5148\u5efa\u7f6e\u5bb9\u5668\uff0c\u4f7f\u5f97\u958b\u767c\u4eba\u54e1\u80fd\u5920\u5c07\u90e8\u7f72\u6642\u9593\u5f9e\u539f\u672c\u7684\u6578\u9031\u7e2e\u77ed\u5230\u5e7e\u5206\u9418\u3002<\/p>\n\n\n\n<p>\u9019\u4e9b\u5fae\u670d\u52d9\u70ba\u8a9e\u8a00\u3001\u8a9e\u97f3\u548c\u85e5\u7269\u767c\u73fe\u7b49\u9818\u57df\u63d0\u4f9b\u4e86\u7b26\u5408\u7522\u696d\u6a19\u6e96\u7684\u61c9\u7528\u7a0b\u5f0f\u4ecb\u9762\uff08API\uff09\uff0c\u8b93\u958b\u767c\u4eba\u54e1\u80fd\u5920\u4f7f\u7528\u5b89\u5168\u8a17\u7ba1\u5728\u81ea\u5df1\u57fa\u790e\u8a2d\u65bd\u4e2d\u7684\u5c08\u6709\u8cc7\u6599\u5feb\u901f\u958b\u767c\u4eba\u5de5\u667a\u6167\u61c9\u7528\u7a0b\u5f0f\u3002\u9019\u4e9b\u61c9\u7528\u7a0b\u5f0f\u53ef\u4ee5\u4f9d\u4f7f\u7528\u9700\u6c42\u8abf\u6574\u898f\u6a21\uff0c\u70ba\u5728 NVIDIA \u52a0\u901f\u904b\u7b97\u5e73\u53f0\u4e0a\u904b\u884c\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u63d0\u4f9b\u4e86\u7d55\u4f73\u7684\u9748\u6d3b\u6027\u548c\u6548\u80fd\u3002<\/p>\n\n\n\n<p>NIM \u5fae\u670d\u52d9\u70ba\u90e8\u7f72 NVIDIA\u3001A121\u3001Adept\u3001Cohere\u3001Getty Images \u548c Shutterstock \u7684\u6a21\u578b\uff0c\u4ee5\u53ca Google\u3001Hugging Face\u3001Meta\u3001Mistral AI \u548c Stability AI \u7684\u958b\u653e\u6a21\u578b\uff0c\u63d0\u4f9b\u4e86\u901f\u5ea6\u6700\u5feb\u3001\u57f7\u884c\u6548\u80fd\u6700\u9ad8\u7684\u751f\u7522\u578b AI \u5bb9\u5668\uff0c\u4e0d\u4e45\u9084\u5c07\u652f\u63f4\u5fae\u8edf\u7684\u6a21\u578b\u3002<\/p>\n\n\n\n<p><a href=\"https:\/\/www.servicenow.com\/company\/media\/press-room\/expand-nvidia-partnership-genai.html\" target=\"_blank\" rel=\"noreferrer noopener\">ServiceNow<\/a> \u4eca\u65e5\u5ba3\u5e03\u8a72\u516c\u53f8\u6b63\u5728\u4f7f\u7528 NIM\uff0c\u4ee5\u66f4\u5feb\u7684\u901f\u5ea6\u8207\u66f4\u9ad8\u7684\u6210\u672c\u6548\u76ca\u958b\u767c\u548c\u90e8\u7f72\u5168\u65b0\u7279\u5b9a\u9818\u57df\u7684AI\u8f14\u52a9\u5925\u4f34\uff0c\u9084\u6709\u5176\u4ed6\u751f\u6210\u5f0f AI \u61c9\u7528\u7a0b\u5f0f\u3002<\/p>\n\n\n\n<p>\u5ba2\u6236\u5c07\u80fd\u5920\u5f9e Amazon SageMaker\u3001Google Kubernetes Engine \u8207 Microsoft Azure AI \u4f7f\u7528 NIM \u5fae\u670d\u52d9\uff0c\u4e26\u4e14\u8207 Deepset\u3001LangChain \u548c LlamaIndex \u7b49\u71b1\u9580\u7684 AI \u6846\u67b6\u9032\u884c\u6574\u5408\u3002<\/p>\n\n\n\n<p><\/p>\n\n\n\n<p><strong>\u7528\u65bcRAG\u3001\u8cc7\u6599\u8655\u7406\u3001Guardrails\u3001\u9ad8\u6548\u80fd\u904b\u7b97\u7684CUDA-X\u5fae\u670d\u52d9<\/strong><\/p>\n\n\n\n<p><a href=\"https:\/\/www.nvidia.com\/en-us\/technologies\/cuda-x\/\" target=\"_blank\" rel=\"noreferrer noopener\">CUDA-X \u5fae\u670d\u52d9<\/a>\u70ba\u8cc7\u6599\u6e96\u5099\u3001\u5ba2\u88fd\u5316\u548c\u8a13\u7df4\u63d0\u4f9b\u7aef\u5230\u7aef\u7684\u5efa\u69cb\u6a21\u7d44\uff0c\u4ee5\u52a0\u5feb\u5404\u884c\u5404\u696d\u958b\u767c\u751f\u7522\u578b\u4eba\u5de5\u667a\u6167\u7684\u901f\u5ea6\u3002<\/p>\n\n\n\n<p>\u4f01\u696d\u53ef\u4ee5\u4f7f\u7528\u591a\u9805 CUDA-X \u5fae\u670d\u52d9\uff0c\u5305\u62ec\u7528\u65bc\u53ef\u81ea\u8a02\u8a9e\u97f3\u548c\u7ffb\u8b6fAI\u7684 <a href=\"https:\/\/www.nvidia.com\/en-us\/ai-data-science\/products\/riva\/\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA Riva<\/a>\u3001\u7528\u65bc\u8a2d\u5b9a\u6700\u4f73\u8def\u5f91\u7684 <a href=\"https:\/\/developer.nvidia.com\/cuopt-logistics-optimization\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA cuOpt<\/a>\u2122\uff0c\u4ee5\u53ca\u7528\u65bc\u9ad8\u89e3\u6790\u5ea6\u6c23\u5019\u548c\u5929\u6c23\u6a21\u64ec\u7684<a href=\"https:\/\/nvidianews.nvidia.com\/news\/nvidia-announces-earth-climate-digital-twin\" target=\"_blank\" rel=\"noreferrer noopener\"> NVIDIA Earth-2<\/a>\uff0c\u4ee5\u52a0\u5feb\u63a1\u7528 AI \u7684\u8173\u6b65\u3002<\/p>\n\n\n\n<p>\u958b\u767c\u4eba\u54e1\u53ef\u4ee5\u900f\u904e <a href=\"https:\/\/developer.nvidia.com\/blog\/translate-your-enterprise-data-into-actionable-insights-with-nvidia-nemo-retriever\/\" target=\"_blank\" rel=\"noreferrer noopener\">NeMo Retriever\u2122<\/a> \u5fae\u670d\u52d9\uff0c\u4e32\u9023 AI \u61c9\u7528\u7a0b\u5f0f\u8207\u6587\u5b57\u3001\u5716\u7247\uff0c\u4ee5\u53ca\u9577\u689d\u5716\u3001\u6298\u7dda\u5716\u548c\u5713\u9905\u5716\u7b49\u8996\u89ba\u5316\u8cc7\u6599\u7b49\u696d\u52d9\u8cc7\u6599\uff0c\u4ee5\u7522\u751f\u51fa\u9ad8\u5ea6\u6e96\u78ba\u53c8\u8207\u8a9e\u5883\u8108\u7d61\u76f8\u95dc\u7684\u56de\u61c9\u3002\u6709\u4e86\u9019\u4e9b RAG \u529f\u80fd\uff0c\u4f01\u696d\u5c31\u80fd\u70ba AI \u8f14\u52a9\u5925\u4f34\u3001\u804a\u5929\u6a5f\u5668\u4eba\u548c\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u751f\u7522\u529b\u5de5\u5177\u63d0\u4f9b\u66f4\u591a\u8cc7\u6599\uff0c\u4ee5\u63d0\u9ad8\u6e96\u78ba\u6027\u548c\u7372\u5f97\u66f4\u5bf6\u8cb4\u7684\u898b\u89e3\u3002<\/p>\n\n\n\n<p>NVIDIA \u4e0d\u4e45\u5f8c\u5c07\u63a8\u51fa\u7528\u65bc\u958b\u767c\u5ba2\u88fd\u5316\u6a21\u578b\u7684\u5176\u4ed6 <a href=\"https:\/\/www.nvidia.com\/en-us\/ai-data-science\/generative-ai\/nemo-framework\/\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA NeMo<\/a>\u2122 \u5fae\u670d\u52d9\u3002\u9019\u4e9b\u670d\u52d9\u5305\u62ec\u91dd\u5c0d\u5efa\u7acb\u7528\u65bc\u8a13\u7df4\u548c\u6aa2\u7d22\u4e4b\u7c21\u6f54\u8cc7\u6599\u96c6\u7684 NVIDIA NeMo Curator\u3001\u7528\u65bc\u5229\u7528\u7279\u5b9a\u9818\u57df\u8cc7\u6599\u5fae\u8abf\u5927\u578b\u8a9e\u8a00\u6a21\u578b \uff08LLM\uff09\u7684 NVIDIA NeMo Customizer\u3001\u7528\u65bc\u5206\u6790\u4eba\u5de5\u667a\u6167\u6a21\u578b\u6548\u80fd\u7684 NVIDIA NeMo Evaluator\uff0c\u4ee5\u53ca\u7528\u65bc LLM \u7684 <a href=\"https:\/\/blogs.nvidia.com\/blog\/ai-chatbot-guardrails-nemo\/\" target=\"_blank\" rel=\"noreferrer noopener\">NVIDIA NeMo Guardrails<\/a>\u3002<\/p>\n\n\n\n<p><\/p>\n\n\n\n<p><strong>\u751f\u614b\u7cfb\u7d71\u5229\u7528\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u5fae\u670d\u52d9\u589e\u5f37\u4f01\u696d\u5e73\u53f0<\/strong><strong><\/strong><\/p>\n\n\n\n<p>\u9664\u4e86\u4e3b\u8981\u7684\u61c9\u7528\u7a0b\u5f0f\u63d0\u4f9b\u5546\uff0c\u6574\u500b NVIDIA \u751f\u614b\u7cfb\u7d71\u4e2d\u7684\u8cc7\u6599\u3001\u57fa\u790e\u67b6\u69cb\u548c\u904b\u7b97\u5e73\u53f0\u63d0\u4f9b\u5546\u540c\u6a23\u8207 NVIDIA \u5fae\u670d\u52d9\u5408\u4f5c\uff0c\u5c07\u751f\u6210\u5f0f AI \u5e36\u7d66\u4f01\u696d\u4f7f\u7528\u3002<\/p>\n\n\n\n<p>\u5305\u62ec Box\u3001Cloudera\u3001Cohesity\u3001Datastax\u3001Dropbox\u3001NetApp \u53ca Snowflake \u5728\u5167\u7684\u9818\u5148\u8cc7\u6599\u5e73\u53f0\u63d0\u4f9b\u5546\u6b63\u5728\u8207 NVIDIA \u5fae\u670d\u52d9\u5408\u4f5c\uff0c\u5354\u52a9\u5ba2\u6236\u5c07\u5176 RAG \u7ba1\u9053\u8abf\u6574\u6700\u4f73\u72c0\u614b\uff0c\u4e26\u5c07\u5176\u5c08\u6709\u8cc7\u6599\u8207\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u61c9\u7528\u7a0b\u5f0f\u9032\u884c\u6574\u5408\u3002Snowflake\u900f\u904eNeMo Retriever\u4f86\u5229\u7528\u4f01\u696d\u6578\u64da\u69cb\u5efa AI \u61c9\u7528\u3002<\/p>\n\n\n\n<p>\u4f01\u696d\u53ef\u4ee5\u5728\u81ea\u5df1\u9078\u64c7\u7684\u57fa\u790e\u67b6\u69cb\u4e0a\u90e8\u7f72 NVIDIA AI Enterprise 5.0 \u6240\u5305\u542b\u7684 NVIDIA \u5fae\u670d\u52d9\uff0c\u4f8b\u5982 Amazon Web Services\uff08AWS\uff09\u3001Google Cloud\u3001Azure \u548c Oracle Cloud Infrastructure \u7b49\u4e3b\u8981\u96f2\u7aef\u904b\u7b97\u5e73\u53f0\u3002<\/p>\n\n\n\n<p>NVIDIA \u5fae\u670d\u52d9\u9084\u652f\u63f4 400 \u591a\u500b 400 NVIDIA \u8a8d\u8b49\u7cfb\u7d71\uff0c\u5305\u62ec\u601d\u79d1\u3001\u6234\u723e\u79d1\u6280\u96c6\u5718\u3001\u6167\u8207\u79d1\u6280\u3001\u60e0\u666e\u3001\u806f\u60f3 \u53ca\u7f8e\u8d85\u5fae\u7b49\u516c\u53f8\u6240\u63a8\u51fa\u7684\u4f3a\u670d\u5668\u548c\u5de5\u4f5c\u7ad9\u3002\u6167\u8207\u79d1\u6280\u4eca\u5929\u53e6\u5ba3\u5e03\u63a8\u51fa\u7528\u65bc\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u7684 \u4f01\u696d\u904b\u7b97\u89e3\u6c7a\u65b9\u6848\uff0c\u4e26\u4e14\u5c07\u628a NIM \u548c NVIDIA AI Foundation \u6a21\u578b\u7d0d\u5165\u6167\u8207\u79d1\u6280\u7684\u4eba\u5de5\u667a\u6167\u8edf\u9ad4\u3002<\/p>\n\n\n\n<p>NVIDIA AI Enterprise \u5fae\u670d\u52d9\u5373\u5c07\u52a0\u5165\u57fa\u790e\u67b6\u69cb\u8edf\u9ad4\u5e73\u53f0\uff0c\u5305\u62ec\u8207 NVIDIA \u5408\u4f5c\u7684 VMware Private AI Foundation\u3002Red Hat OpenShift \u652f\u63f4 NVIDIA NIM \u5fae\u670d\u52d9\uff0c\u5354\u52a9\u4f01\u696d\u66f4\u8f15\u9b06\u5730\u5c07\u751f\u6210\u5f0f\u4eba\u5de5\u667a\u6167\u529f\u80fd\u8207\u61c9\u7528\u7a0b\u5f0f\u9032\u884c\u6574\u5408\uff0c\u4e26\u4e14\u7372\u5f97\u6700\u9ad8\u7684\u5b89\u5168\u6027\u3001\u5408\u898f\u6027\u548c\u63a7\u5236\u529f\u80fd\u3002Canonical \u6b63\u900f\u904e NVIDIA AI Enterprise \u70ba NVIDIA \u5fae\u670d\u52d9\u52a0\u5165\u652f\u63f4 Charmed Kubernetes \u7684\u80fd\u529b\u3002<\/p>\n\n\n\n<p>NVIDIA \u751f\u614b\u7cfb\u7d71\u7531\u6578\u767e\u5bb6\u4eba\u5de5\u667a\u6167\u548c MLOps \u5408\u4f5c\u5925\u4f34\u7d44\u6210\uff0c\u5176\u4e2d\u5305\u62ec Abridge\u3001Anyscale\u3001Dataiku\u3001DataRobot\u3001Glean\u3001H2O.ai\u3001Securiti AI\u3001Scale.ai\u3001OctoAI \u53ca Weights &amp; Biases\uff0c\u5b83\u5011\u5c07\u900f\u904e NVIDIA AI Enterprise \u4f86\u52a0\u5165\u652f\u63f4 NVIDIA \u5fae\u670d\u52d9\u3002<\/p>\n\n\n\n<p>Apache Lucene\u3001Datastax\u3001Faiss\u3001Kinetica\u3001Milvus\u3001Redis \u548c Weaviate \u7b49\u5411\u91cf\u641c\u5c0b\u63d0\u4f9b\u5546\u8207 NVIDIA NeMo Retriever \u5fae\u670d\u52d9\u5408\u4f5c\uff0c\u70ba\u4f01\u696d\u63d0\u4f9b\u53cd\u61c9\u9748\u654f\u7684 RAG \u529f\u80fd\u3002<\/p>\n\n\n\n<p><\/p>\n\n\n\n<p><strong>\u4f9b\u8ca8\u6642\u7a0b<\/strong><strong><\/strong><\/p>\n\n\n\n<p>\u958b\u767c\u4eba\u54e1\u53ef\u4ee5\u5728 <a href=\"https:\/\/ai.nvidia.com\" target=\"_blank\" rel=\"noreferrer noopener\">ai.nvidia.com<\/a> \u7db2\u7ad9\u514d\u8cbb\u8a66\u7528 NVIDIA \u5fae\u670d\u52d9\u3002\u4f01\u696d\u53ef\u4ee5\u5229\u7528\u5728 NVIDIA \u8a8d\u8b49\u7cfb\u7d71\u53ca\u4e3b\u8981\u96f2\u7aef\u5e73\u53f0\u4e0a\u904b\u884c\u7684 NVIDIA AI Enterprise 5.0 \u90e8\u7f72\u751f\u7522\u7d1a NIM \u5fae\u670d\u52d9\u3002<\/p>\n\n\n\n<p>\u6b32\u77ad\u89e3\u66f4\u591a\u8cc7\u8a0a\uff0c\u8acb\u89c0\u770b<a href=\"https:\/\/www.nvidia.com\/gtc\/keynote\/\" target=\"_blank\" rel=\"noreferrer noopener\">\u9ec3\u4ec1\u52f3\u65bcGTC\u7684\u4e3b\u984c\u6f14\u8b1b<\/a>\uff0c\u4e26\u53c3\u89c0 NVIDIA 3 \u6708 18 \u81f3 21 \u65e5\u5728 GTC \u7684\u73fe\u5834\u6524\u4f4d\u3002<\/p>\n\n\n\n<p><\/p>\n\n\n\n<p><\/p>\n\n\n\n<h4 class=\"wp-block-heading has-text-align-right has-very-light-gray-to-cyan-bluish-gray-gradient-background has-background\">\ud83d\udfe6<a href=\"https:\/\/reurl.cc\/rrG1r1\" target=\"_blank\" rel=\"noreferrer noopener\">\u770bbe quiet! Dark Rock Pro 5\u6587\u7ae0 \u62bdSilent Wings 4 PWM\u98a8\u6247<\/a><br>\ud83d\udfe6<a href=\"https:\/\/reurl.cc\/prb1d4\" target=\"_blank\" rel=\"noopener\">\u56de\u7b54 Archer BE55 \u901a\u95dc\u79d8\u8a9e \u9001\u60a8\u7db2\u8def\u651d\u5f71\u6a5f<\/a><br>\ud83d\udfe6<strong>\u73fe\u5728\u5c31\u52a0\u5165&nbsp;<a href=\"https:\/\/www.facebook.com\/profile.php?id=100086628162118\" target=\"_blank\" rel=\"noreferrer noopener\">ioioTIMES \u81c9\u66f8\u7c89\u7d72\u5718<\/a>&nbsp;\u66f4\u591a\u4e92\u52d5\u3001\u66f4\u591a\u597d\u5eb7\u650f\u62b5\u52a0!!<\/strong><br>\ud83d\udfe6<strong>\u6211\u5011\u6709<a href=\"https:\/\/today.line.me\/tw\/v2\/publisher\/103117\" target=\"_blank\" rel=\"noreferrer noopener nofollow\">LINE TODAY<\/a>\u983b\u9053\u4e86\uff0c\u5feb\u4f86\u8ffd\u8e2a\u6211\u5011\u5427!!&#8211;\u6700\u65b0\u79d1\u6280\u65b0\u805e \u76e1\u5728\u4f60\u624b<\/strong><\/h4>\n","protected":false},"excerpt":{"rendered":"<p>NVIDIA \u4eca\u65e5\u63a8\u51fa\u6578\u5341\u7a2e\u4f01\u696d\u7d1a\u751f<\/p>\n","protected":false},"author":3,"featured_media":62332,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"rank_math_lock_modified_date":false,"footnotes":""},"categories":[13],"tags":[71,9070,9071,9069],"class_list":["post-62016","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-news","tag-nvidia","tag-nvidia-cuda-gpu","tag-nvidia-nim","tag-9069"],"_links":{"self":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/62016"}],"collection":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/users\/3"}],"replies":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcomments&post=62016"}],"version-history":[{"count":3,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/62016\/revisions"}],"predecessor-version":[{"id":62334,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/posts\/62016\/revisions\/62334"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=\/wp\/v2\/media\/62332"}],"wp:attachment":[{"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fmedia&parent=62016"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Fcategories&post=62016"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.ioiotimes.com\/index.php?rest_route=%2Fwp%2Fv2%2Ftags&post=62016"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}