{"id":34899,"date":"2025-08-04T10:29:58","date_gmt":"2025-08-04T02:29:58","guid":{"rendered":"https:\/\/www.kdjingpai.com\/?p=34899"},"modified":"2025-08-04T10:29:58","modified_gmt":"2025-08-04T02:29:58","slug":"lmcache","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/en\/lmcache\/","title":{"rendered":"LMCache\uff1a\u52a0\u901f\u5927\u8bed\u8a00\u6a21\u578b\u63a8\u7406\u7684\u952e\u503c\u7f13\u5b58\u4f18\u5316\u5de5\u5177"},"content":{"rendered":"<p>LMCache \u662f\u4e00\u4e2a\u5f00\u6e90\u7684\u952e\u503c\uff08KV\uff09\u7f13\u5b58\u4f18\u5316\u5de5\u5177\uff0c\u4e13\u4e3a\u63d0\u5347\u5927\u8bed\u8a00\u6a21\u578b\uff08LLM\uff09\u63a8\u7406\u6548\u7387\u8bbe\u8ba1\u3002\u5b83\u901a\u8fc7\u7f13\u5b58\u548c\u590d\u7528\u6a21\u578b\u7684\u4e2d\u95f4\u8ba1\u7b97\u7ed3\u679c\uff08\u952e\u503c\u7f13\u5b58\uff09\uff0c\u663e\u8457\u964d\u4f4e\u63a8\u7406\u65f6\u95f4\u548cGPU\u8d44\u6e90\u6d88\u8017\uff0c\u7279\u522b\u9002\u5408\u957f\u4e0a\u4e0b\u6587\u573a\u666f\u3002LMCache \u4e0e <a href=\"https:\/\/www.kdjingpai.com\/de\/vllm\/\">vLLM<\/a> \u7b49\u63a8\u7406\u5f15\u64ce\u65e0\u7f1d\u96c6\u6210\uff0c\u652f\u6301 GPU\u3001CPU \u548c\u78c1\u76d8\u5b58\u50a8\uff0c\u9002\u7528\u4e8e\u591a\u8f6e\u95ee\u7b54\u3001\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09\u7b49\u573a\u666f\u3002\u9879\u76ee\u7531\u793e\u533a\u9a71\u52a8\uff0c\u91c7\u7528 Apache 2.0 \u8bb8\u53ef\u8bc1\uff0c\u5e7f\u6cdb\u5e94\u7528\u4e8e\u4f01\u4e1a\u7ea7 AI \u63a8\u7406\u4f18\u5316\u3002<\/p>\n<p><img decoding=\"async\" class=\"aligncenter\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/08\/ce0fafc4793f045-scaled.png\" alt=\"\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li><strong>\u952e\u503c\u7f13\u5b58\u590d\u7528<\/strong>\uff1a\u7f13\u5b58 LLM \u7684\u952e\u503c\u5bf9\uff0c\u652f\u6301\u975e\u524d\u7f00\u6587\u672c\u590d\u7528\uff0c\u51cf\u5c11\u91cd\u590d\u8ba1\u7b97\u3002<\/li>\n<li><strong>\u591a\u5b58\u50a8\u540e\u7aef\u652f\u6301<\/strong>\uff1a\u652f\u6301 GPU\u3001CPU DRAM\u3001\u78c1\u76d8\u548c Redis \u7b49\u5b58\u50a8\uff0c\u7075\u6d3b\u5e94\u5bf9\u5185\u5b58\u9650\u5236\u3002<\/li>\n<li><strong>\u4e0e vLLM \u96c6\u6210<\/strong>\uff1a\u65e0\u7f1d\u63a5\u5165 vLLM\uff0c\u63d0\u4f9b 3-10 \u500d\u7684\u63a8\u7406\u5ef6\u8fdf\u4f18\u5316\u3002<\/li>\n<li><strong>\u5206\u5e03\u5f0f\u7f13\u5b58<\/strong>\uff1a\u652f\u6301\u8de8\u591a GPU \u6216\u5bb9\u5668\u5316\u73af\u5883\u5171\u4eab\u7f13\u5b58\uff0c\u9002\u5408\u5927\u89c4\u6a21\u90e8\u7f72\u3002<\/li>\n<li><strong>\u591a\u6a21\u6001\u652f\u6301<\/strong>\uff1a\u7f13\u5b58\u56fe\u50cf\u548c\u6587\u672c\u7684\u952e\u503c\u5bf9\uff0c\u4f18\u5316\u591a\u6a21\u6001\u6a21\u578b\u63a8\u7406\u3002<\/li>\n<li><strong>\u5de5\u4f5c\u8d1f\u8f7d\u751f\u6210<\/strong>\uff1a\u63d0\u4f9b\u6d4b\u8bd5\u5de5\u5177\u751f\u6210\u591a\u8f6e\u95ee\u7b54\u3001RAG \u7b49\u5de5\u4f5c\u8d1f\u8f7d\uff0c\u9a8c\u8bc1\u6027\u80fd\u3002<\/li>\n<li><strong>\u5f00\u6e90\u793e\u533a\u652f\u6301<\/strong>\uff1a\u63d0\u4f9b\u6587\u6863\u3001\u793a\u4f8b\u548c\u793e\u533a\u4f1a\u8bae\uff0c\u4fbf\u4e8e\u7528\u6237\u8d21\u732e\u548c\u4ea4\u6d41\u3002<\/li>\n<\/ul>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<p>LMCache \u7684\u5b89\u88c5\u7b80\u5355\uff0c\u652f\u6301 Linux \u5e73\u53f0\u548c NVIDIA GPU \u73af\u5883\u3002\u4ee5\u4e0b\u662f\u8be6\u7ec6\u7684\u5b89\u88c5\u6b65\u9aa4\uff0c\u57fa\u4e8e\u5b98\u65b9\u6587\u6863\u548c\u793e\u533a\u63a8\u8350\u7684\u65b9\u5f0f\u3002<\/p>\n<ol>\n<li><strong>\u51c6\u5907\u73af\u5883<\/strong>\uff1a\n<ul>\n<li>\u786e\u4fdd\u7cfb\u7edf\u4e3a Linux\uff0cPython \u7248\u672c\u4e3a 3.10 \u6216\u66f4\u9ad8\uff0cCUDA \u7248\u672c\u4e3a 12.1 \u6216\u4ee5\u4e0a\u3002<\/li>\n<li>\u5b89\u88c5 Conda\uff08\u63a8\u8350\u4f7f\u7528 Miniconda\uff09\u4ee5\u521b\u5efa\u865a\u62df\u73af\u5883\uff1a\n<pre><code>conda create -n lmcache python=3.10\r\nconda activate lmcache\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u514b\u9686\u4ed3\u5e93<\/strong>\uff1a\n<ul>\n<li>\u4f7f\u7528 Git \u514b\u9686 LMCache \u4ed3\u5e93\u5230\u672c\u5730\uff1a\n<pre><code>git clone https:\/\/github.com\/LMCache\/LMCache.git\r\ncd LMCache\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u5b89\u88c5 LMCache<\/strong>\uff1a\n<ul>\n<li>\u901a\u8fc7 PyPI \u5b89\u88c5\u6700\u65b0\u7a33\u5b9a\u7248\uff1a\n<pre><code>pip install lmcache\r\n<\/code><\/pre>\n<\/li>\n<li>\u6216\u8005\u5b89\u88c5\u6700\u65b0\u9884\u53d1\u5e03\u7248\uff08\u53ef\u80fd\u5305\u542b\u5b9e\u9a8c\u6027\u529f\u80fd\uff09\uff1a\n<pre><code>pip install --index-url https:\/\/pypi.org\/simple --extra-index-url https:\/\/test.pypi.org\/simple lmcache==0.2.2.dev57\r\n<\/code><\/pre>\n<\/li>\n<li>\u82e5\u9700\u4ece\u6e90\u7801\u5b89\u88c5\uff1a\n<pre><code>pip install -e .\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u5b89\u88c5 vLLM<\/strong>\uff1a\n<ul>\n<li>LMCache \u9700\u4e0e vLLM \u914d\u5408\u4f7f\u7528\uff0c\u5b89\u88c5\u6700\u65b0\u7248 vLLM\uff1a\n<pre><code>pip install vllm\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u9a8c\u8bc1\u5b89\u88c5<\/strong>\uff1a\n<ul>\n<li>\u68c0\u67e5 LMCache \u662f\u5426\u6b63\u786e\u5b89\u88c5\uff1a\n<pre><code>python\r\nimport lmcache\r\nfrom importlib.metadata import version\r\nprint(version(\"lmcache\"))\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\u5e94\u4e3a\u5b89\u88c5\u7684\u7248\u672c\u53f7\uff0c\u5982\u00a0<code>0.2.2.dev57<\/code>\u3002<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u53ef\u9009\uff1aDocker \u90e8\u7f72<\/strong>\uff1a\n<ul>\n<li>LMCache \u63d0\u4f9b\u9884\u6784\u5efa\u7684 Docker \u955c\u50cf\uff0c\u96c6\u6210 vLLM\uff1a\n<pre><code>docker pull lmcache\/lmcache:latest\r\n<\/code><\/pre>\n<\/li>\n<li>\u8fd0\u884c Docker \u5bb9\u5668\u5e76\u6839\u636e\u6587\u6863\u914d\u7f6e vLLM \u548c LMCache\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u4f7f\u7528\u4e3b\u8981\u529f\u80fd<\/h3>\n<p>LMCache \u7684\u6838\u5fc3\u529f\u80fd\u662f\u4f18\u5316\u952e\u503c\u7f13\u5b58\uff0c\u4ee5\u52a0\u901f LLM \u63a8\u7406\u3002\u4ee5\u4e0b\u662f\u4e3b\u8981\u529f\u80fd\u7684\u8be6\u7ec6\u64cd\u4f5c\u6307\u5357\u3002<\/p>\n<h4>1. \u952e\u503c\u7f13\u5b58\u590d\u7528<\/h4>\n<p>LMCache \u901a\u8fc7\u5b58\u50a8\u6a21\u578b\u7684\u952e\u503c\u7f13\u5b58\uff08KV Cache\uff09\uff0c\u907f\u514d\u91cd\u590d\u8ba1\u7b97\u76f8\u540c\u7684\u6587\u672c\u6216\u4e0a\u4e0b\u6587\u3002\u7528\u6237\u53ef\u4ee5\u5728 vLLM \u4e2d\u542f\u7528 LMCache\uff1a<\/p>\n<ul>\n<li><strong>\u914d\u7f6e\u73af\u5883\u53d8\u91cf<\/strong>\uff1a\n<pre><code>export LMCACHE_USE_EXPERIMENTAL=True\r\nexport LMCACHE_CHUNK_SIZE=256\r\nexport LMCACHE_LOCAL_CPU=True\r\nexport LMCACHE_MAX_LOCAL_CPU_SIZE=5.0\r\n<\/code><\/pre>\n<p>\u8fd9\u4e9b\u53d8\u91cf\u8bbe\u7f6e LMCache \u4f7f\u7528\u5b9e\u9a8c\u529f\u80fd\u3001\u6bcf\u5757 256 \u4e2a token\u3001\u542f\u7528 CPU \u540e\u7aef\uff0c\u5e76\u9650\u5236 CPU \u5185\u5b58\u4e3a 5GB\u3002<\/li>\n<li><strong>\u8fd0\u884c vLLM \u5b9e\u4f8b<\/strong>\uff1a<br \/>\n\u542f\u52a8 vLLM \u65f6\uff0cLMCache \u81ea\u52a8\u52a0\u8f7d\u5e76\u7f13\u5b58\u952e\u503c\u5bf9\u3002\u793a\u4f8b\u4ee3\u7801\uff1a<\/p>\n<pre><code>from vllm import LLM\r\nfrom lmcache.integration.vllm.utils import ENGINE_NAME\r\nfrom vllm.config import KVTransferConfig\r\nktc = KVTransferConfig(kv_connector=\"LMCacheConnector\", kv_role=\"kv_both\")\r\nllm = LLM(model=\"meta-llama\/Meta-Llama-3.1-8B-Instruct\", kv_transfer_config=ktc)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h4>2. \u591a\u5b58\u50a8\u540e\u7aef<\/h4>\n<p>LMCache \u652f\u6301\u5c06\u952e\u503c\u7f13\u5b58\u5b58\u50a8\u5728 GPU\u3001CPU\u3001\u78c1\u76d8\u6216 Redis \u4e0a\u3002\u7528\u6237\u53ef\u6839\u636e\u786c\u4ef6\u8d44\u6e90\u9009\u62e9\u5b58\u50a8\u65b9\u5f0f\uff1a<\/p>\n<ul>\n<li><strong>\u672c\u5730\u78c1\u76d8\u5b58\u50a8<\/strong>\uff1a\n<pre><code>python3 -m lmcache_server.server localhost 9000 \/path\/to\/disk\r\n<\/code><\/pre>\n<p>\u8fd9\u5c06\u542f\u52a8 LMCache \u670d\u52a1\u5668\uff0c\u5b58\u50a8\u7f13\u5b58\u5230\u6307\u5b9a\u78c1\u76d8\u8def\u5f84\u3002<\/li>\n<li><strong>Redis \u5b58\u50a8<\/strong>\uff1a<br \/>\n\u914d\u7f6e Redis \u540e\u7aef\u9700\u8bbe\u7f6e\u7528\u6237\u540d\u548c\u5bc6\u7801\uff0c\u53c2\u8003\u6587\u6863\uff1a<\/p>\n<pre><code>export LMCACHE_REDIS_USERNAME=user\r\nexport LMCACHE_REDIS_PASSWORD=pass\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<h4>3. \u5206\u5e03\u5f0f\u7f13\u5b58<\/h4>\n<p>\u5728\u591a GPU \u6216\u5bb9\u5668\u5316\u73af\u5883\u4e2d\uff0cLMCache \u652f\u6301\u8de8\u8282\u70b9\u5171\u4eab\u7f13\u5b58\uff1a<\/p>\n<ul>\n<li>\u542f\u52a8 LMCache \u670d\u52a1\u5668\uff1a\n<pre><code>python3 -m lmcache_server.server localhost 9000 cpu\r\n<\/code><\/pre>\n<\/li>\n<li>\u914d\u7f6e vLLM \u5b9e\u4f8b\u8fde\u63a5\u670d\u52a1\u5668\uff0c\u53c2\u8003\u00a0<code>disagg_vllm_launcher.sh<\/code>\u00a0\u793a\u4f8b\u3002<\/li>\n<\/ul>\n<h4>4. \u591a\u6a21\u6001\u652f\u6301<\/h4>\n<p>LMCache \u652f\u6301\u591a\u6a21\u6001\u6a21\u578b\uff0c\u901a\u8fc7\u54c8\u5e0c\u56fe\u50cf token\uff08mm_hashes\uff09\u7f13\u5b58\u952e\u503c\u5bf9\uff0c\u4f18\u5316\u89c6\u89c9 LLM \u63a8\u7406\uff1a<\/p>\n<ul>\n<li>\u5728 vLLM \u4e2d\u542f\u7528\u591a\u6a21\u6001\u652f\u6301\uff0c\u53c2\u8003\u5b98\u65b9\u793a\u4f8b\u00a0<code>LMCache-Examples<\/code>\u00a0\u4ed3\u5e93\u3002<\/li>\n<\/ul>\n<h4>5. \u6d4b\u8bd5\u5de5\u5177<\/h4>\n<p>LMCache \u63d0\u4f9b\u6d4b\u8bd5\u5de5\u5177\uff0c\u751f\u6210\u5de5\u4f5c\u8d1f\u8f7d\u4ee5\u9a8c\u8bc1\u6027\u80fd\uff1a<\/p>\n<ul>\n<li>\u514b\u9686\u6d4b\u8bd5\u4ed3\u5e93\uff1a\n<pre><code>git clone https:\/\/github.com\/LMCache\/lmcache-tests.git\r\ncd lmcache-tests\r\nbash prepare_environment.sh\r\n<\/code><\/pre>\n<\/li>\n<li>\u8fd0\u884c\u6d4b\u8bd5\u7528\u4f8b\uff1a\n<pre><code>python3 main.py tests\/tests.py -f test_lmcache_local_cpu -o outputs\/\r\n<\/code><\/pre>\n<p>\u8f93\u51fa\u7ed3\u679c\u4fdd\u5b58\u5728\u00a0<code>outputs\/test_lmcache_local_cpu.csv<\/code>\u3002<\/li>\n<\/ul>\n<h3>\u64cd\u4f5c\u6ce8\u610f\u4e8b\u9879<\/h3>\n<ul>\n<li><strong>\u73af\u5883\u68c0\u67e5<\/strong>\uff1a\u786e\u4fdd CUDA \u548c Python \u7248\u672c\u517c\u5bb9\uff0c\u63a8\u8350\u4f7f\u7528 Conda \u7ba1\u7406\u73af\u5883\u3002<\/li>\n<li><strong>\u65e5\u5fd7\u76d1\u63a7<\/strong>\uff1a\u68c0\u67e5\u00a0<code>prefiller.log<\/code>\u3001<code>decoder.log<\/code>\u00a0\u548c\u00a0<code>proxy.log<\/code>\u00a0\u4ee5\u8c03\u8bd5\u95ee\u9898\u3002<\/li>\n<li><strong>\u793e\u533a\u652f\u6301<\/strong>\uff1a\u52a0\u5165 LMCache Slack \u6216\u53c2\u52a0 bi-weekly \u793e\u533a\u4f1a\u8bae\u83b7\u53d6\u5e2e\u52a9\uff0c\u4f1a\u8bae\u65f6\u95f4\u4e3a\u6bcf\u5468\u4e8c\u4e0a\u5348 9 \u70b9\uff08PT\uff09\u3002<\/li>\n<\/ul>\n<h2>\u5e94\u7528\u573a\u666f<\/h2>\n<ol>\n<li><strong>\u591a\u8f6e\u95ee\u7b54\u7cfb\u7edf<\/strong><br \/>\nLMCache \u7f13\u5b58\u4e0a\u4e0b\u6587\u4e2d\u7684\u952e\u503c\u5bf9\uff0c\u52a0\u901f\u591a\u8f6e\u5bf9\u8bdd\u573a\u666f\u3002\u7528\u6237\u5728\u804a\u5929\u673a\u5668\u4eba\u4e2d\u8fde\u7eed\u63d0\u95ee\u65f6\uff0cLMCache \u590d\u7528\u4e4b\u524d\u7684\u8ba1\u7b97\u7ed3\u679c\uff0c\u51cf\u5c11\u5ef6\u8fdf\u3002<\/li>\n<li><strong>\u68c0\u7d22\u589e\u5f3a\u751f\u6210\uff08RAG\uff09<\/strong><br \/>\n\u5728 <a href=\"https:\/\/www.kdjingpai.com\/de\/rag\/\">RAG<\/a> \u5e94\u7528\u4e2d\uff0cLMCache \u7f13\u5b58\u6587\u6863\u7684\u952e\u503c\u5bf9\uff0c\u5feb\u901f\u54cd\u5e94\u76f8\u4f3c\u67e5\u8be2\uff0c\u9002\u5408\u667a\u80fd\u6587\u6863\u641c\u7d22\u6216\u4f01\u4e1a\u77e5\u8bc6\u5e93\u3002<\/li>\n<li><strong>\u591a\u6a21\u6001\u6a21\u578b\u63a8\u7406<\/strong><br \/>\n\u5bf9\u4e8e\u89c6\u89c9-\u8bed\u8a00\u6a21\u578b\uff0cLMCache \u7f13\u5b58\u56fe\u50cf\u548c\u6587\u672c\u7684\u952e\u503c\u5bf9\uff0c\u964d\u4f4e GPU \u5185\u5b58\u5360\u7528\uff0c\u63d0\u5347\u54cd\u5e94\u901f\u5ea6\u3002<\/li>\n<li><strong>\u5927\u89c4\u6a21\u5206\u5e03\u5f0f\u90e8\u7f72<\/strong><br \/>\n\u5728\u591a GPU \u6216\u5bb9\u5668\u5316\u73af\u5883\u4e2d\uff0cLMCache \u7684\u5206\u5e03\u5f0f\u7f13\u5b58\u529f\u80fd\u652f\u6301\u8de8\u8282\u70b9\u5171\u4eab\uff0c\u4f18\u5316\u4f01\u4e1a\u7ea7 AI \u63a8\u7406\u3002<\/li>\n<\/ol>\n<h2>QA<\/h2>\n<ol>\n<li><strong>LMCache \u652f\u6301\u54ea\u4e9b\u5e73\u53f0\uff1f<\/strong><br \/>\n\u76ee\u524d\u652f\u6301 Linux \u548c NVIDIA GPU \u73af\u5883\uff0cWindows \u53ef\u901a\u8fc7 WSL \u4f7f\u7528\u3002<\/li>\n<li><strong>\u5982\u4f55\u4e0e vLLM \u96c6\u6210\uff1f<\/strong><br \/>\n\u901a\u8fc7\u00a0<code>pip install lmcache vllm<\/code>\uff0c\u5e76\u5728 vLLM \u914d\u7f6e\u4e2d\u542f\u7528 LMCacheConnector\uff0c\u53c2\u8003\u5b98\u65b9\u793a\u4f8b\u4ee3\u7801\u3002<\/li>\n<li><strong>\u662f\u5426\u652f\u6301\u975e\u524d\u7f00\u7f13\u5b58\uff1f<\/strong><br \/>\n\u652f\u6301\uff0cLMCache \u4f7f\u7528\u90e8\u5206\u91cd\u8ba1\u7b97\u6280\u672f\uff0c\u7f13\u5b58 RAG \u5de5\u4f5c\u8d1f\u8f7d\u4e2d\u7684\u975e\u524d\u7f00\u6587\u672c\u3002<\/li>\n<li><strong>\u5982\u4f55\u8c03\u8bd5\u6027\u80fd\u95ee\u9898\uff1f<\/strong><br \/>\n\u68c0\u67e5\u65e5\u5fd7\u6587\u4ef6\u5e76\u8fd0\u884c\u6d4b\u8bd5\u7528\u4f8b\uff0c\u8f93\u51fa CSV \u6587\u4ef6\u5206\u6790\u5ef6\u8fdf\u548c\u541e\u5410\u91cf\u3002<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>LMCache \u662f\u4e00\u4e2a\u5f00\u6e90\u7684\u952e\u503c\uff08KV\uff09\u7f13\u5b58\u4f18\u5316\u5de5\u5177\uff0c\u4e13\u4e3a\u63d0\u5347\u5927\u8bed\u8a00\u6a21\u578b\uff08LLM\uff09\u63a8\u7406\u6548\u7387\u8bbe\u8ba1\u3002\u5b83\u901a\u8fc7\u7f13\u5b58\u548c\u590d\u7528\u6a21\u578b\u7684\u4e2d\u95f4\u8ba1\u7b97\u7ed3\u679c\uff08\u952e\u503c\u7f13\u5b58\uff09\uff0c\u663e\u8457\u964d\u4f4e\u63a8\u7406\u65f6\u95f4\u548cGPU\u8d44\u6e90\u6d88\u8017\uff0c\u7279\u522b\u9002\u5408\u957f\u4e0a\u4e0b\u6587\u573a\u666f\u3002LMCache \u4e0e vLLM \u7b49\u63a8\u7406\u5f15\u64ce&#8230;<\/p>\n","protected":false},"author":1,"featured_media":32782,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[425,20,497],"tags":[232],"class_list":["post-34899","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-ai-professional","category-tool","category-deployment-model","tag-bendebushukaiyuanba"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/34899","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/comments?post=34899"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/34899\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media\/32782"}],"wp:attachment":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/media?parent=34899"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/categories?post=34899"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/tags?post=34899"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}