{"id":6574,"date":"2025-11-04T11:12:00","date_gmt":"2025-11-04T11:12:00","guid":{"rendered":"https:\/\/lp.szlogic.cn\/glossary\/tpu-tensor-processing-unit-google-ai-accelerator\/"},"modified":"2026-06-22T05:34:25","modified_gmt":"2026-06-22T05:34:25","slug":"tpu-tensor-processing-unit-google-ai-accelerator","status":"publish","type":"post","link":"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/tpu-tensor-processing-unit-google-ai-accelerator","title":{"rendered":"Memahami TPU: Di Dalam Arsitektur Tensor Processing Unit Google"},"content":{"rendered":"<figure class=\"wp-block-image aligncenter size-large\"><img fetchpriority=\"high\" decoding=\"async\" width=\"1200\" height=\"712\" src=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259.webp\" alt=\"What Is TPU?\" class=\"wp-image-6570\" srcset=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259.webp 1200w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259-300x178.webp 300w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259-1024x608.webp 1024w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259-768x456.webp 768w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/54d67b7b0d92483599dd22af221ec259-18x12.webp 18w\" sizes=\"(max-width: 1200px) 100vw, 1200px\" \/><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Apa Itu TPU (Tensor Processing Unit)?<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">A <strong>Tensor Processing Unit (TPU)<\/strong> adalah akselerator AI khusus yang dikembangkan oleh Google untuk mempercepat beban kerja pembelajaran mesin\u2014terutama operasi pembelajaran mendalam yang dibangun di atas komputasi tensor dan matriks skala besar. Berbeda dengan CPU atau GPU, TPU dirancang khusus <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/what-is-application-specific-integrated-circuit-asic\/\">ASIC<\/a> untuk pelatihan dan inferensi jaringan saraf berkinerja tinggi dan efisien energi dalam skala besar.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Mengapa Google Membuat TPU<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\" >Dioptimalkan untuk Pembelajaran Mendalam<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Jaringan saraf membutuhkan operasi matematika paralel dalam jumlah besar, terutama tugas perkalian-akumulasi matriks. <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/what-is-cpu-central-processing-unit\/\"><strong>CPU<\/strong><\/a> kesulitan menjalankan beban kerja ini, sedangkan <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/what-is-a-gpu-graphics-processing-units\/\"><strong>GPU<\/strong><\/a>, meskipun kuat, merupakan akselerator serba guna.<br\/><strong>TPU <\/strong>dibuat untuk:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><p>Memberikan kinerja per watt yang sangat tinggi<\/p><\/li><li><p>Memaksimalkan throughput perkalian matriks<\/p><\/li><li><p>Mendukung model AI skala besar secara hemat biaya<\/p><\/li><li><p>Memenuhi permintaan internal yang meningkat di seluruh Google Search, Translate, YouTube, Maps, dan model AI<\/p><\/li>\n<\/ul>\n\n\n\n<h3 class=\"wp-block-heading\" >Desain Berbasis AI<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Sejak awal, <strong>arsitektur TPU<\/strong> berfokus pada:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><p>Ko-desain perangkat keras-perangkat lunak dengan TensorFlow<\/p><\/li><li><p>Format presisi tereduksi (misalnya bfloat16, int8) untuk komputasi hemat energi<\/p><\/li><li><p>Fabrik yang dapat diskalakan untuk pengelompokan multi-chip<\/p><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Penjelasan Arsitektur TPU<\/h2>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\"><img decoding=\"async\" width=\"1536\" height=\"1024\" src=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e.png\" alt=\"TPU Architecture\" class=\"wp-image-6571\" srcset=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e.png 1536w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e-300x200.png 300w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e-1024x683.png 1024w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e-768x512.png 768w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/d1ac50e745d64fcb9d389c7931db629e-18x12.png 18w\" sizes=\"(max-width: 1536px) 100vw, 1536px\" \/><\/figure>\n\n\n\n<h3 class=\"wp-block-heading\" >Mesin Matriks Sistolik<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Di inti setiap chip TPU terdapat <strong>unit perkalian matriks berukuran besar<\/strong> yang disusun dalam susunan sistolik, memungkinkan ribuan operasi perkalian-akumulasi secara bersamaan.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" >Memori Berbandwidth Tinggi<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">TPU modern mengintegrasikan <strong>HBM<\/strong> untuk mengalirkan data pada bandwidth sangat tinggi, mencegah bottleneck memori yang umum terjadi pada sistem berbasis GPU.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" >Interkoneksi &amp; Skalabilitas<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">TPU individu dapat diskalakan menjadi <strong>TPU Pod<\/strong>, yang saling terhubung melalui jaringan berlatensi rendah dan bandwidth tinggi guna membentuk klaster AI modular multi-exaflop.<br\/>Arsitektur ini memungkinkan pelatihan model berukuran sangat besar dan inferensi lebih cepat pada skala hiperskala.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Generasi TPU dan Spesifikasi Utama<\/h2>\n\n\n\n<figure class=\"wp-block-table\">\n<table class=\"has-fixed-layout\">\n<colgroup><col style=\"width: 134px;\"\/><col style=\"width: 200px;\"\/><col style=\"width: 179px;\"\/><col style=\"min-width: 25px;\"\/><\/colgroup><tbody><tr><th colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Generasi<\/p><\/th><th colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Focus<\/p><\/th><th colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>Memori &amp; Komputasi<\/p><\/th><th colspan=\"1\" rowspan=\"1\"><p>Catatan<\/p><\/th><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v1<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Inferensi<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>Komputasi 8-bit<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Penyebaran internal pertama<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v2<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Pelatihan &amp; Inferensi<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>bfloat16, HBM<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Peluncuran Cloud TPU<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v3<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Pelatihan skala besar<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>Pendinginan cair, HBM<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Pod hingga ~1.000 chip<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v4<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Pod eksaskala yang efisien<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>HBM 32 GB, mesh canggih<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Skala pusat data<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v6 \u201cTrillium\u201d<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Komputasi AI berkepadatan tinggi<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>Beberapa tumpukan HBM<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>~5\u00d7 peningkatan kinerja dibanding versi sebelumnya<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>TPU v7 \u201cIronwood\u201d<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"200\"><p>Arsitektur yang mengutamakan inferensi<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"179\"><p>Optimisasi FP8<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Dibangun khusus untuk pelayanan LLM<\/p><\/td><\/tr><\/tbody>\n<\/table>\n<\/figure>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f TPU vs GPU vs CPU<\/h2>\n\n\n\n<figure class=\"wp-block-image aligncenter size-large\"><img decoding=\"async\" width=\"1200\" height=\"315\" src=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99.webp\" alt=\"TPU vs GPU vs CPU\" class=\"wp-image-6572\" srcset=\"https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99.webp 1200w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99-300x79.webp 300w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99-1024x269.webp 1024w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99-768x202.webp 768w, https:\/\/resourceslp.szlogic.cn\/wp-content\/uploads\/2026\/05\/a83e15692e184550860b10ac91d93a99-18x5.webp 18w\" sizes=\"(max-width: 1200px) 100vw, 1200px\" \/><\/figure>\n\n\n\n<figure class=\"wp-block-table\">\n<table class=\"has-fixed-layout\">\n<colgroup><col style=\"width: 134px;\"\/><col style=\"width: 194px;\"\/><col style=\"min-width: 25px;\"\/><col style=\"min-width: 25px;\"\/><\/colgroup><tbody><tr><th colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Fitur<\/p><\/th><th colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>TPU<\/p><\/th><th colspan=\"1\" rowspan=\"1\"><p><a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/what-is-a-gpu-graphics-processing-units\/\">GPU<\/a><\/p><\/th><th colspan=\"1\" rowspan=\"1\"><p><a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/glossary\/what-is-cpu-central-processing-unit\/\">CPU<\/a><\/p><\/th><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Tujuan<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>Komputasi tensor khusus AI<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Akselerasi grafis + ML<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Komputasi umum<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Paling Cocok Untuk<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>Jaringan saraf dan LLM<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>HPC, ML, grafis<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>OS, logika, aplikasi<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Paralelisme<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>Sangat tinggi<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>High<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Low<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Efisiensi<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>Tertinggi untuk beban kerja AI<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>High<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Tujuan umum<\/p><\/td><\/tr><tr><td colspan=\"1\" rowspan=\"1\" colwidth=\"134\"><p>Penyebaran<\/p><\/td><td colspan=\"1\" rowspan=\"1\" colwidth=\"194\"><p>Cloud &amp; klaster<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Cloud &amp; on-prem<\/p><\/td><td colspan=\"1\" rowspan=\"1\"><p>Di mana-mana<\/p><\/td><\/tr><\/tbody>\n<\/table>\n<\/figure>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>Singkatnya:<\/strong><\/p>\n\n\n\n<blockquote class=\"wp-block-quote is-layout-flow wp-block-quote-is-layout-flow\"><p><em>CPU bersifat universal. GPU serba guna. TPU sangat terfokus pada AI dalam skala besar.<\/em><\/p><\/blockquote>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Di Mana TPU Digunakan<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\" >Pelatihan Model Berskala Besar<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Ideal untuk model transformer, sistem rekomendasi, dan saluran pelatihan model bahasa besar.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" >Inferensi Cloud<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">TPU menggerakkan beban kerja AI global <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/knowledge-center\/link-pp-optical-modules-ai-iot-big-data-performance-reliability\/\">beban kerja AI<\/a> seperti perankingan pencarian, penerjemahan bahasa, pengenalan ucapan, dan layanan AI generatif.<\/p>\n\n\n\n<h3 class=\"wp-block-heading\" >Edge TPU<\/h3>\n\n\n\n<p class=\"wp-block-paragraph\">Varian TPU ringan yang menjalankan inferensi ML secara lokal di perangkat tepi\/tertanam untuk kecerdasan AI berlatensi rendah dan hemat daya <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/knowledge-center\/iot-internet-of-things-definition-and-real-world-examples\/\">IoT<\/a> kecerdasan.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Praktik Terbaik untuk Penyebaran TPU<\/h2>\n\n\n\n<ul class=\"wp-block-list\">\n<li><p>Gunakan jenis data yang didukung (bfloat16 \/ int8) untuk efisiensi maksimal<\/p><\/li><li><p>Optimalkan saluran data untuk komputasi terdistribusi<\/p><\/li><li><p>Pilih TPU Pod untuk beban kerja berskala LLM<\/p><\/li><li><p>Pertimbangkan desain termal dan jaringan untuk skalabilitas klaster<\/p><\/li><li><p>Manfaatkan strategi hybrid cloud + edge guna mencapai kepadatan komputasi yang seimbang<\/p><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f TPUs dan Masa Depan Infrastruktur AI<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Model AI kini lebih intensif komputasi daripada sebelumnya, menggeser fokus dari pelatihan murni ke <strong>inferensi waktu nyata dalam skala besar<\/strong>.<br\/>TPU akan terus berkembang dalam hal:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><p>Kepadatan interkoneksi<\/p><\/li><li><p>Arsitektur hemat energi<\/p><\/li><li><p>Presisi hibrida (misalnya, FP8)<\/p><\/li><li><p>Integrasi dengan kerangka kerja perangkat lunak (TensorFlow, JAX, PyTorch melalui XLA)<\/p><\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Seiring percepatan beban kerja AI, komputasi khusus dan konektivitas berkecepatan ultra-tinggi menjadi komponen penting dalam <a target=\"_blank\" rel=\"\" href=\"https:\/\/resourceslp.szlogic.cn\/id\/knowledge-center\/what-is-a-data-center\/\">desain pusat data modern<\/a> dan jaringan.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Hubungan Ini dengan LINK-PP<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Akselerasi AI dalam skala hiperskala bergantung pada jaringan canggih dan infrastruktur konektivitas yang andal. <a target=\"_blank\" rel=\"\" href=\"https:\/\/www.l-p.com\/\">LINK-PP<\/a> komponen mendukung lingkungan pusat data yang menggerakkan penyebaran TPU, termasuk:<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><p>Lingkungan penyimpanan kritis misi dengan latensi rendah <a target=\"_blank\" rel=\"\" href=\"https:\/\/www.l-p.com\/store-17492-integrated-rj45-connector.htm\"><strong>MagJacks RJ45<\/strong><\/a><\/p><\/li><li><p><strong>SFP\/25G\/100G<\/strong> <a target=\"_blank\" rel=\"\" href=\"https:\/\/www.l-p.com\/store-25432-optics-transceivers-sfp-modules.htm\">modul optik<\/a><\/p><\/li><li><p><strong>PoE<\/strong> solusi untuk perangkat AI tepi<\/p><\/li><li><p>Konektor Ethernet Industri &amp; IoT<\/p><\/li>\n<\/ul>\n\n\n\n<h2 class=\"wp-block-heading\" >\u2699\ufe0f Kesimpulan<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\"><strong>TPU<\/strong> mewakili lompatan besar dalam komputasi AI khusus <strong>AI komputasi<\/strong>\u2014yang dirancang khusus untuk beban kerja tensor dan operasi jaringan saraf skala besar. Seiring percepatan adopsi AI generatif dan pembelajaran mendalam secara global, TPU memainkan peran penting dalam menggerakkan kluster pelatihan dan infrastruktur inferensi.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">Bagi industri yang membangun atau mendukung lingkungan pusat data modern, memahami teknologi TPU memberikan wawasan berharga mengenai tuntutan sistem AI berkinerja tinggi\u2014serta peluang dalam perangkat keras dan komponen jaringan generasi berikutnya.<\/p>","protected":false},"excerpt":{"rendered":"<p>Pelajari apa itu TPU\u2014Tensor Processing Unit, cara kerja akselerator AI Google, generasi-generasi utama TPU, perbandingan TPU versus GPU, serta perannya dalam pembelajaran mesin skala besar yang efisien.<\/p>","protected":false},"author":1,"featured_media":6573,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_acf_changed":false,"footnotes":""},"categories":[27],"tags":[22,24,26],"class_list":["post-6574","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-glossary","tag-integrated-rj45-connectors","tag-link-pp","tag-optics-transceivers"],"blocksy_meta":[],"acf":[],"_links":{"self":[{"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/posts\/6574","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/comments?post=6574"}],"version-history":[{"count":5,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/posts\/6574\/revisions"}],"predecessor-version":[{"id":10935,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/posts\/6574\/revisions\/10935"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/media\/6573"}],"wp:attachment":[{"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/media?parent=6574"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/categories?post=6574"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/resourceslp.szlogic.cn\/id\/wp-json\/wp\/v2\/tags?post=6574"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}