{"id":29023,"date":"2025-03-18T21:53:04","date_gmt":"2025-03-18T13:53:04","guid":{"rendered":"https:\/\/www.aisharenet.com\/?p=29023"},"modified":"2025-08-25T00:09:57","modified_gmt":"2025-08-24T16:09:57","slug":"smoldocling","status":"publish","type":"post","link":"https:\/\/www.kdjingpai.com\/en\/smoldocling\/","title":{"rendered":"SmolDocling\uff1a\u5c0f\u4f53\u79ef\u9ad8\u6548\u5904\u7406\u6587\u6863\u7684\u89c6\u89c9\u8bed\u8a00\u6a21\u578b"},"content":{"rendered":"<p>SmolDocling \u662f\u7531 ds4sd \u56e2\u961f\u4e0e IBM \u5408\u4f5c\u5f00\u53d1\u7684\u4e00\u4e2a\u89c6\u89c9\u8bed\u8a00\u6a21\u578b\uff08VLM\uff09\uff0c\u57fa\u4e8e SmolVLM-256M \u6253\u9020\uff0c\u6258\u7ba1\u5728 Hugging Face \u5e73\u53f0\u3002\u5b83\u4f53\u79ef\u5c0f\uff0c\u53ea\u6709 256M \u53c2\u6570\uff0c\u5374\u662f\u5168\u7403\u6700\u5c0f\u7684 VLM\u3002\u5b83\u7684\u6838\u5fc3\u529f\u80fd\u662f\u4ece\u56fe\u7247\u4e2d\u63d0\u53d6\u6587\u5b57\u3001\u8bc6\u522b\u5e03\u5c40\u3001\u4ee3\u7801\u3001\u516c\u5f0f\u548c\u56fe\u8868\uff0c\u5e76\u751f\u6210\u7ed3\u6784\u5316\u7684 DocTags \u683c\u5f0f\u6587\u6863\u3002SmolDocling \u5728\u666e\u901a\u8bbe\u5907\u4e0a\u5c31\u80fd\u8fd0\u884c\uff0c\u6548\u7387\u9ad8\uff0c\u8d44\u6e90\u5360\u7528\u5c11\u3002\u5f00\u53d1\u56e2\u961f\u901a\u8fc7\u5f00\u6e90\u65b9\u5f0f\u5206\u4eab\u8fd9\u4e2a\u6a21\u578b\uff0c\u5e0c\u671b\u5e2e\u52a9\u66f4\u591a\u4eba\u5904\u7406\u6587\u6863\u4efb\u52a1\u3002\u5b83\u662f SmolVLM \u5bb6\u65cf\u7684\u4e00\u90e8\u5206\uff0c\u4e13\u6ce8\u4e8e\u6587\u6863\u8f6c\u6362\uff0c\u9002\u5408\u9700\u8981\u5feb\u901f\u5904\u7406\u590d\u6742\u6587\u6863\u7684\u7528\u6237\u3002<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-29024\" title=\"SmolDocling\uff1a\u5c0f\u4f53\u79ef\u9ad8\u6548\u5904\u7406\u6587\u6863\u7684\u89c6\u89c9\u8bed\u8a00\u6a21\u578b-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b.jpg\" alt=\"SmolDocling\uff1a\u5c0f\u4f53\u79ef\u9ad8\u6548\u5904\u7406\u6587\u6863\u7684\u89c6\u89c9\u8bed\u8a00\u6a21\u578b-1\" width=\"2500\" height=\"1600\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b.jpg 2500w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b-768x492.jpg 768w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b-1536x983.jpg 1536w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b-2048x1311.jpg 2048w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/9bb19742ff2fa9b-18x12.jpg 18w\" sizes=\"auto, (max-width: 2500px) 100vw, 2500px\" \/><\/p>\n<p>&nbsp;<\/p>\n<p><img loading=\"lazy\" decoding=\"async\" class=\"aligncenter size-full wp-image-29025\" title=\"SmolDocling\uff1a\u5c0f\u4f53\u79ef\u9ad8\u6548\u5904\u7406\u6587\u6863\u7684\u89c6\u89c9\u8bed\u8a00\u6a21\u578b-1\" src=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/d698e70bc250428.png\" alt=\"SmolDocling\uff1a\u5c0f\u4f53\u79ef\u9ad8\u6548\u5904\u7406\u6587\u6863\u7684\u89c6\u89c9\u8bed\u8a00\u6a21\u578b-1\" width=\"707\" height=\"577\" srcset=\"https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/d698e70bc250428.png 707w, https:\/\/www.kdjingpai.com\/wp-content\/uploads\/2025\/03\/d698e70bc250428-15x12.png 15w\" sizes=\"auto, (max-width: 707px) 100vw, 707px\" \/><\/p>\n<p>&nbsp;<\/p>\n<h2>\u529f\u80fd\u5217\u8868<\/h2>\n<ul>\n<li><strong>\u6587\u5b57\u63d0\u53d6\uff08OCR\uff09<\/strong>\uff1a\u4ece\u56fe\u7247\u4e2d\u8bc6\u522b\u5e76\u63d0\u53d6\u6587\u5b57\uff0c\u652f\u6301\u591a\u8bed\u8a00\u3002<\/li>\n<li><strong>\u5e03\u5c40\u8bc6\u522b<\/strong>\uff1a\u5206\u6790\u56fe\u7247\u4e2d\u6587\u6863\u7684\u7ed3\u6784\uff0c\u5982\u6807\u9898\u3001\u6bb5\u843d\u3001\u8868\u683c\u7684\u4f4d\u7f6e\u3002<\/li>\n<li><strong>\u4ee3\u7801\u8bc6\u522b<\/strong>\uff1a\u63d0\u53d6\u4ee3\u7801\u5757\u5e76\u4fdd\u7559\u7f29\u8fdb\u548c\u683c\u5f0f\u3002<\/li>\n<li><strong>\u516c\u5f0f\u8bc6\u522b<\/strong>\uff1a\u68c0\u6d4b\u6570\u5b66\u516c\u5f0f\u5e76\u8f6c\u4e3a\u53ef\u7f16\u8f91\u6587\u672c\u3002<\/li>\n<li><strong>\u56fe\u8868\u8bc6\u522b<\/strong>\uff1a\u89e3\u6790\u56fe\u7247\u4e2d\u7684\u56fe\u8868\u5185\u5bb9\u5e76\u63d0\u53d6\u6570\u636e\u3002<\/li>\n<li><strong>\u8868\u683c\u5904\u7406<\/strong>\uff1a\u8bc6\u522b\u8868\u683c\u7ed3\u6784\uff0c\u4fdd\u7559\u884c\u5217\u4fe1\u606f\u3002<\/li>\n<li><strong>DocTags \u8f93\u51fa<\/strong>\uff1a\u5c06\u5904\u7406\u7ed3\u679c\u8f6c\u4e3a\u7edf\u4e00\u7684\u6807\u8bb0\u683c\u5f0f\uff0c\u65b9\u4fbf\u540e\u7eed\u4f7f\u7528\u3002<\/li>\n<li><strong>\u9ad8\u5206\u8fa8\u7387\u56fe\u50cf\u5904\u7406<\/strong>\uff1a\u652f\u6301\u66f4\u5927\u5206\u8fa8\u7387\u7684\u56fe\u7247\u8f93\u5165\uff0c\u63d0\u5347\u8bc6\u522b\u7cbe\u5ea6\u3002<\/li>\n<\/ul>\n<p>&nbsp;<\/p>\n<h2>\u4f7f\u7528\u5e2e\u52a9<\/h2>\n<p>SmolDocling \u7684\u4f7f\u7528\u5206\u4e3a\u5b89\u88c5\u548c\u64cd\u4f5c\u4e24\u90e8\u5206\u3002\u4ee5\u4e0b\u662f\u8be6\u7ec6\u6b65\u9aa4\uff0c\u5e2e\u52a9\u7528\u6237\u5feb\u901f\u4e0a\u624b\u3002<\/p>\n<h3>\u5b89\u88c5\u6d41\u7a0b<\/h3>\n<ol>\n<li><strong>\u51c6\u5907\u73af\u5883<\/strong>\n<ul>\n<li>\u786e\u4fdd\u7535\u8111\u5df2\u5b89\u88c5 Python 3.8 \u6216\u66f4\u9ad8\u7248\u672c\u3002<\/li>\n<li>\u5728\u7ec8\u7aef\u8f93\u5165\u4ee5\u4e0b\u547d\u4ee4\u5b89\u88c5\u4f9d\u8d56\u5e93\uff1a\n<pre><code>pip install torch transformers docling_core\r\n<\/code><\/pre>\n<\/li>\n<li>\u5982\u679c\u6709 GPU\uff0c\u63a8\u8350\u5b89\u88c5\u652f\u6301 CUDA \u7684 PyTorch\uff0c\u8fd0\u884c\u66f4\u5feb\u3002\u68c0\u67e5\u65b9\u6cd5\uff1a\n<pre><code>import torch\r\nprint(\"GPU\u53ef\u7528\uff1a\" if torch.cuda.is_available() else \"\u4f7f\u7528CPU\")\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u52a0\u8f7d\u6a21\u578b<\/strong>\n<ul>\n<li>SmolDocling \u4e0d\u9700\u8981\u624b\u52a8\u4e0b\u8f7d\uff0c\u76f4\u63a5\u901a\u8fc7\u4ee3\u7801\u4ece Hugging Face \u83b7\u53d6\u3002<\/li>\n<li>\u786e\u4fdd\u7f51\u7edc\u7545\u901a\uff0c\u9996\u6b21\u8fd0\u884c\u4f1a\u81ea\u52a8\u4e0b\u8f7d\u6a21\u578b\u6587\u4ef6\u3002<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u4f7f\u7528\u6b65\u9aa4<\/h3>\n<ol>\n<li><strong>\u51c6\u5907\u56fe\u7247<\/strong>\n<ul>\n<li>\u627e\u4e00\u5f20\u5305\u542b\u6587\u5b57\u7684\u56fe\u7247\uff0c\u6bd4\u5982\u626b\u63cf\u6587\u6863\u6216\u622a\u56fe\u3002<\/li>\n<li>\u7528\u4ee3\u7801\u52a0\u8f7d\u56fe\u7247\uff1a\n<pre><code>from transformers.image_utils import load_image\r\nimage = load_image(\"\u4f60\u7684\u56fe\u7247\u8def\u5f84.jpg\")\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u521d\u59cb\u5316\u6a21\u578b\u548c\u5904\u7406\u5668<\/strong>\n<ul>\n<li>\u52a0\u8f7d SmolDocling \u7684\u5904\u7406\u5668\u548c\u6a21\u578b\uff1a\n<pre><code>from transformers import AutoProcessor, AutoModelForVision2Seq\r\nDEVICE = \"cuda\" if torch.cuda.is_available() else \"cpu\"\r\nprocessor = AutoProcessor.from_pretrained(\"ds4sd\/SmolDocling-256M-preview\")\r\nmodel = AutoModelForVision2Seq.from_pretrained(\r\n\"ds4sd\/SmolDocling-256M-preview\",\r\ntorch_dtype=torch.bfloat16\r\n).to(DEVICE)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u751f\u6210 DocTags<\/strong>\n<ul>\n<li>\u8bbe\u7f6e\u8f93\u5165\u5e76\u8fd0\u884c\u6a21\u578b\uff1a\n<pre><code>messages = [{\"role\": \"user\", \"content\": [{\"type\": \"image\"}, {\"type\": \"text\", \"text\": \"Convert this page to docling.\"}]}]\r\nprompt = processor.apply_chat_template(messages, add_generation_prompt=True)\r\ninputs = processor(text=prompt, images=[image], return_tensors=\"pt\").to(DEVICE)\r\ngenerated_ids = model.generate(**inputs, max_new_tokens=8192)\r\ndoctags = processor.batch_decode(generated_ids[:, inputs.input_ids.shape[1]:], skip_special_tokens=False)[0].lstrip()\r\nprint(doctags)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u8f6c\u4e3a\u5e38\u7528\u683c\u5f0f<\/strong>\n<ul>\n<li>\u5c06 DocTags \u8f6c\u4e3a Markdown \u6216\u5176\u4ed6\u683c\u5f0f\uff1a\n<pre><code>from docling_core.types.doc import DoclingDocument\r\ndoc = DoclingDocument(name=\"\u6211\u7684\u6587\u6863\")\r\ndoc.load_from_doctags(doctags)\r\nprint(doc.export_to_markdown())\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<li><strong>\u9ad8\u7ea7\u7528\u6cd5\uff08\u53ef\u9009\uff09<\/strong>\n<ul>\n<li><strong>\u5904\u7406\u591a\u9875\u6587\u6863<\/strong>\uff1a\u7528\u5faa\u73af\u5904\u7406\u591a\u5f20\u56fe\u7247\uff0c\u7136\u540e\u5408\u5e76 DocTags\u3002<\/li>\n<li><strong>\u4f18\u5316\u6027\u80fd<\/strong>\uff1a\u8bbe\u7f6e\u00a0<code>torch_dtype=torch.bfloat16<\/code>\u00a0\u8282\u7701\u5185\u5b58\uff0cGPU \u7528\u6237\u53ef\u542f\u7528\u00a0<code>flash_attention_2<\/code>\u00a0\u52a0\u901f\uff1a\n<pre><code>model = AutoModelForVision2Seq.from_pretrained(\r\n\"ds4sd\/SmolDocling-256M-preview\",\r\ntorch_dtype=torch.bfloat16,\r\n_attn_implementation=\"flash_attention_2\" if DEVICE == \"cuda\" else \"eager\"\r\n).to(DEVICE)\r\n<\/code><\/pre>\n<\/li>\n<\/ul>\n<\/li>\n<\/ol>\n<h3>\u64cd\u4f5c\u6280\u5de7<\/h3>\n<ul>\n<li><strong>\u56fe\u7247\u8981\u6c42<\/strong>\uff1a\u56fe\u7247\u9700\u6e05\u6670\uff0c\u6587\u5b57\u53ef\u8fa8\u8bc6\uff0c\u5206\u8fa8\u7387\u8d8a\u9ad8\u6548\u679c\u8d8a\u597d\u3002<\/li>\n<li><strong>\u8c03\u6574\u53c2\u6570<\/strong>\uff1a\u5982\u679c\u7ed3\u679c\u4e0d\u5b8c\u6574\uff0c\u589e\u52a0\u00a0<code>max_new_tokens<\/code>\uff08\u9ed8\u8ba4 8192\uff09\u3002<\/li>\n<li><strong>\u6279\u91cf\u5904\u7406<\/strong>\uff1a\u591a\u5f20\u56fe\u7247\u53ef\u4ee5\u7528\u5217\u8868\u4f20\u5165\u00a0<code>images=[image1, image2]<\/code>\u3002<\/li>\n<li><strong>\u8c03\u8bd5\u65b9\u6cd5<\/strong>\uff1a\u8f93\u51fa\u4e2d\u95f4\u7ed3\u679c\u68c0\u67e5\uff0c\u6bd4\u5982\u6253\u5370\u00a0<code>inputs<\/code>\u00a0\u67e5\u770b\u8f93\u5165\u662f\u5426\u6b63\u786e\u3002<\/li>\n<\/ul>\n<h3>\u6ce8\u610f\u4e8b\u9879<\/h3>\n<ul>\n<li>\u9996\u6b21\u8fd0\u884c\u9700\u8054\u7f51\uff0c\u4e4b\u540e\u53ef\u79bb\u7ebf\u4f7f\u7528\u3002<\/li>\n<li>\u56fe\u7247\u8fc7\u5927\u53ef\u80fd\u5bfc\u81f4\u5185\u5b58\u4e0d\u8db3\uff0c\u5efa\u8bae\u88c1\u526a\u540e\u5904\u7406\u3002<\/li>\n<li>\u5982\u679c\u9047\u5230\u9519\u8bef\uff0c\u68c0\u67e5 Python \u7248\u672c\u548c\u4f9d\u8d56\u5e93\u662f\u5426\u6b63\u786e\u5b89\u88c5\u3002<\/li>\n<\/ul>\n<p>\u901a\u8fc7\u4ee5\u4e0a\u6b65\u9aa4\uff0c\u7528\u6237\u53ef\u4ee5\u7528 SmolDocling \u628a\u56fe\u7247\u8f6c\u4e3a\u7ed3\u6784\u5316\u6587\u6863\u3002\u6574\u4e2a\u8fc7\u7a0b\u7b80\u5355\uff0c\u9002\u5408\u521d\u5b66\u8005\u548c\u4e13\u4e1a\u7528\u6237\u3002<\/p>\n<p>&nbsp;<\/p>\n<h2>\u5e94\u7528\u573a\u666f<\/h2>\n<ol>\n<li><strong>\u5b66\u672f\u7814\u7a76<\/strong><br \/>\n\u5c06\u626b\u63cf\u7684\u8bba\u6587\u8f6c\u4e3a\u6587\u672c\uff0c\u63d0\u53d6\u516c\u5f0f\u548c\u8868\u683c\uff0c\u65b9\u4fbf\u7f16\u8f91\u548c\u5f15\u7528\u3002<\/li>\n<li><strong>\u7f16\u7a0b\u6587\u6863\u6574\u7406<\/strong><br \/>\n\u628a\u5305\u542b\u4ee3\u7801\u7684\u624b\u518c\u56fe\u7247\u8f6c\u4e3a Markdown\uff0c\u4fdd\u7559\u4ee3\u7801\u683c\u5f0f\uff0c\u9002\u5408\u5f00\u53d1\u8005\u4f7f\u7528\u3002<\/li>\n<li><strong>\u529e\u516c\u81ea\u52a8\u5316<\/strong><br \/>\n\u5904\u7406\u5408\u540c\u3001\u62a5\u544a\u7b49\u626b\u63cf\u4ef6\uff0c\u8bc6\u522b\u5e03\u5c40\u548c\u5185\u5bb9\uff0c\u63d0\u9ad8\u5de5\u4f5c\u6548\u7387\u3002<\/li>\n<li><strong>\u6559\u80b2\u652f\u6301<\/strong><br \/>\n\u5c06\u6559\u6750\u56fe\u7247\u8f6c\u4e3a\u53ef\u7f16\u8f91\u6587\u6863\uff0c\u5e2e\u52a9\u6559\u5e08\u548c\u5b66\u751f\u6574\u7406\u7b14\u8bb0\u3002<\/li>\n<\/ol>\n<p>&nbsp;<\/p>\n<h2>QA<\/h2>\n<ol>\n<li><strong>SmolDocling \u548c SmolVLM \u6709\u4ec0\u4e48\u533a\u522b\uff1f<\/strong><br \/>\nSmolDocling \u662f\u57fa\u4e8e SmolVLM-256M \u4f18\u5316\u7684\u7248\u672c\uff0c\u4e13\u6ce8\u4e8e\u6587\u6863\u5904\u7406\uff0c\u8f93\u51fa DocTags \u683c\u5f0f\uff0c\u800c SmolVLM \u66f4\u901a\u7528\uff0c\u652f\u6301\u56fe\u50cf\u63cf\u8ff0\u7b49\u4efb\u52a1\u3002<\/li>\n<li><strong>\u652f\u6301\u54ea\u4e9b\u64cd\u4f5c\u7cfb\u7edf\uff1f<\/strong><br \/>\n\u652f\u6301 Windows\u3001Mac \u548c Linux\uff0c\u53ea\u8981\u5b89\u88c5 Python \u548c\u4f9d\u8d56\u5e93\u5373\u53ef\u8fd0\u884c\u3002<\/li>\n<li><strong>\u5904\u7406\u901f\u5ea6\u5feb\u5417\uff1f<\/strong><br \/>\n\u5728\u666e\u901a\u7535\u8111\u4e0a\u5904\u7406\u4e00\u5f20\u56fe\u7247\u53ea\u9700\u51e0\u79d2\uff0cGPU \u7528\u6237\u66f4\u5feb\uff0c\u901a\u5e38\u4e0d\u5230 1 \u79d2\u3002<\/li>\n<li><strong>\u53ef\u4ee5\u5904\u7406\u624b\u5199\u6587\u5b57\u5417\uff1f<\/strong><br \/>\n\u53ef\u4ee5\uff0c\u4f46\u6548\u679c\u53d6\u51b3\u4e8e\u5b57\u8ff9\u6e05\u6670\u5ea6\uff0c\u5efa\u8bae\u4f7f\u7528\u5370\u5237\u6587\u5b57\u56fe\u7247\u4ee5\u83b7\u5f97\u6700\u4f73\u7ed3\u679c\u3002<\/li>\n<\/ol>\n","protected":false},"excerpt":{"rendered":"<p>SmolDocling \u662f\u7531 ds4sd \u56e2\u961f\u4e0e IBM \u5408\u4f5c\u5f00\u53d1\u7684\u4e00\u4e2a\u89c6\u89c9\u8bed\u8a00\u6a21\u578b\uff08VLM\uff09\uff0c\u57fa\u4e8e SmolVLM-256M \u6253\u9020\uff0c\u6258\u7ba1\u5728 Hugging Face \u5e73\u53f0\u3002\u5b83\u4f53\u79ef\u5c0f\uff0c\u53ea\u6709 256M \u53c2\u6570\uff0c\u5374\u662f\u5168\u7403\u6700\u5c0f\u7684 VLM\u3002\u5b83\u7684\u6838\u5fc3\u529f&#8230;<\/p>\n","protected":false},"author":1,"featured_media":32782,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[20,499],"tags":[230,248,252],"class_list":["post-29023","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-tool","category-document-extraction","tag-aikaiyuanxiangmu","tag-ocr","tag-markdown"],"_links":{"self":[{"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/29023","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=29023"}],"version-history":[{"count":0,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/posts\/29023\/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=29023"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/categories?post=29023"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.kdjingpai.com\/en\/wp-json\/wp\/v2\/tags?post=29023"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}