圖像工作台

模型

提示詞2,710 / 1,000

{"type":"中文科技產業鏈資訊圖海報","theme":"未來 TPU 產業鏈對比：谷歌生態 vs 英偉達生態","canvas":"16:9 寬屏，深海藍色未來感資料中心背景，發光藍色電路走線，高對比度，精緻商務科技展示風格","headline":"[圍繞 TPU 的未來產業鏈：谷歌生態 vs 英偉達生態]","subheadline":"[谷歌自研 TPU + 博通/聯發科 + 臺積電 + 中際旭創，打造超越英偉達的技術未來]","top_row":{"description":"五個圓角卡片透過白色箭頭從左至右相連，代表谷歌 TPU 供應鏈。","count":5,"cards":[{"position":"far left","title":"Google（谷歌）","subtitle":"生態核心/需求方","visual":"Google 標誌，藍色圖示，帶藍色光暈的白色卡片","bullets":["自研晶片架構：TPU","專為AI訓練/推理最佳化","高效能 低功耗 大規模叢集","AI 雲 & 模型驅動","Gemini / Search / YouTube / Cloud","算力需求爆發的核心驅動力"]},{"position":"left center","title":"BROADCOM","subtitle":"高階定製晶片設計","visual":"Broadcom 標誌與黑色 ASIC 晶片渲染圖","bullets":["負責 TPU 核心 ASIC 設計（訓練晶片為主）","高效能運算架構","高速 SerDes / 網路晶片","先進封裝設計能力"]},{"position":"center","title":"MEDIATEK 聯發科","subtitle":"晶片設計（協同+補充）","visual":"聯發科標誌與黑色小型晶片渲染圖","bullets":["負責 TPU I/O、外圍、推理晶片（成本最佳化 + 多樣化）","I/O 晶片 / 介面","推理晶片 / 邊緣AI","成本最佳化專家"]},{"position":"right center","title":"tsmc 臺積電","subtitle":"全球最先進晶片製造","visual":"橙色矽晶圓上的臺積電標誌","bullets":["製造所有 TPU 晶片（3nm / 2nm 領先）","全球最先進製程","最強產能 & 良率","AI時代的“水電煤”"]},{"position":"far right","title":"中際旭創","subtitle":"高速光模組供應","visual":"藍色光模組外形","bullets":["提供 800G / 1.6T 高速光模組","連線 TPU 叢集","光通訊龍頭","AI算力的“高速公路”","資料傳輸的核心保障"]}]},"centerpiece":{"title":"谷歌 TPU 叢集（訓練 & 推理）","visual":"居中前景為大型發光 TPU 晶片，兩側為兩排黑色伺服器機櫃，藍色霓虹平臺，頂部供應商卡片的箭頭匯聚指向 TPU 叢集","labels":["TPU"]},"side_boxes":{"count":2,"boxes":[{"position":"middle left","title":"benefit notes","style":"藍色半透明圓角矩形，帶火箭、葉子和金幣圖示","items":["更高效能：專為AI最佳化，效率更高","更低功耗：TPU 架構天然節能","更低成本：自研+協同設計，成本可控"]},{"position":"middle right","title":"核心優勢","style":"藍色半透明圓角矩形，帶獎盃圖示和綠色對勾","items":["自研架構，擺脫對手技術限制","多方協同，成本更低，迭代更快","完整生態，雲-晶片-網路-光模組閉環","規模效應，長期超越英偉達生態"]}]},"bottom_row":{"description":"對標英偉達生態對比條，包含五個由箭頭從左至右連線的綠色卡片，以及一個紅色劣勢卡片。","comparison_label":"對標：英偉達生態","count":6,"cards":[{"position":"far left","title":"英偉達（NVIDIA）","subtitle":"GPU 架構設計","visual":"英偉達標誌與 GPU 顯示卡","notes":["通用GPU","CUDA生態"]},{"position":"left center","title":"臺積電（TSMC）","subtitle":"晶片製造","visual":"矽晶圓","notes":["同樣依賴臺積電"]},{"position":"center left","title":"臺積電封裝 / 代工夥伴","subtitle":"封裝","visual":"含晶片的先進封裝基板","notes":["封裝產能受限"]},{"position":"center right","title":"光模組供應商（多家）","subtitle":"光通訊連線","visual":"光模組線纜","notes":["需求旺盛","供不應求"]},{"position":"right center","title":"客戶/雲廠","subtitle":"","visual":"微軟、Meta 和 AWS 標誌","notes":["依賴英偉達GPU","成本高，供應緊張"]},{"position":"far right","title":"Nvidia ecosystem disadvantages","style":"紅色描邊卡片，帶紅色叉號圖示","items":["通用架構，效率不如專用","功耗高，成本高","供應鏈單一，受限多","生態封閉，客戶依賴強"]}]},"footer_banner":"[未來格局：谷歌 TPU 生態 = 更高效 + 更低成本 + 更可控 → 有望超越英偉達，成為AI時代新王者！]","style_details":"清晰中文排版，粗體白色標題其中 TPU 為橙色，谷歌生態文字為藍色，英偉達生態文字為綠色，圓角微光卡片，3D 晶片渲染圖，發光連線線條，深藍賽博科技美學，文字清晰可讀，無人物出現。"}

Ctrl / ⌘ + Enter

參考0 / 16

畫幅

品質

解析度

圖像數量

1 / 4

結果

從一句提示詞開始

生成的圖像將出現在此。

獲得靈感