{"id":10596,"date":"2026-08-01T01:52:07","date_gmt":"2026-07-31T17:52:07","guid":{"rendered":"https:\/\/infernews.com\/blog\/spatialcli\/"},"modified":"2026-08-01T01:52:07","modified_gmt":"2026-07-31T17:52:07","slug":"spatialcli","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/spatialcli\/","title":{"rendered":"SpatialCLI \u7528\u7a7a\u9593\u5de5\u5177\u88dc\u5f37\u8996\u89ba\u63a8\u7406"},"content":{"rendered":"\n<figure class=\"wp-block-image size-large\"><img decoding=\"async\" src=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2026\/08\/pasted-c9a29339658e.jpg\" alt=\"A comparison between a general VLM and a general VLM augmented with SpatialCLI tools\"><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">\u4e00\u5230\u8981\u6307\u51fa\u7269\u4ef6\u4f4d\u7f6e\u3001\u5206\u6e05\u906e\u64cb\u95dc\u4fc2\uff0c\u6216\u8005\u4f30\u8a08\u6df1\u5ea6\u8207\u59ff\u614b\uff0c\u7d14 Vision-Language Model \u5f80\u5f80\u6703\u7b54\u5230\u6709\u65b9\u5411\u4f46\u672a\u5920\u6e96\u3002SpatialCLI \u628a\u5462\u500b\u843d\u5dee\u8655\u7406\u5f97\u5e7e\u76f4\u63a5\uff1a\u5b83\u4e0d\u662f\u55ae\u9760\u4e00\u500b\u5927\u6a21\u578b\u786c\u6490\uff0c\u800c\u662f\u628a\u7a7a\u9593\u80fd\u529b\u62c6\u958b\uff0c\u5148\u8b93\u6a21\u578b\u61c2\u5f97\u547c\u53eb\u5de5\u5177\uff0c\u518d\u9032\u4e00\u6b65\u628a\u9019\u4e9b\u80fd\u529b\u5b78\u56de\u81ea\u5df1\u8eab\u4e0a\uff1b\u6574\u9ad4\u5b9a\u4f4d\u66f4\u50cf\u4e00\u500b\u7d50\u5408\u6a21\u578b\u3001\u5de5\u5177\u93c8\u8207\u8a13\u7df4\u65b9\u6cd5\u7684\u7814\u7a76\u9805\u76ee\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u5b83\u6700\u6709\u610f\u601d\u7684\u5730\u65b9\uff0c\u5728\u65bc\u4e09\u6bb5\u5f0f Call-Learn-Internalize\u3002\u7b2c\u4e00\u6b65\u5148\u63a5\u4e0a\u505a localization\u3001segmentation\u3001metric depth\u3001pose \u7684 specialist vision models\uff0c\u7b2c\u4e8c\u6b65\u7528 Cold-Start SFT \u8207 agentic RL \u8a13\u7df4\u6a21\u578b\u5224\u65b7\u5e7e\u6642\u8981\u7528\u54ea\u500b\u5de5\u5177\u3001\u600e\u6a23\u6574\u5408\u7d50\u679c\uff0c\u7b2c\u4e09\u6b65\u518d\u628a\u6210\u529f\u8ecc\u8de1\u8f49\u56de\u6a21\u578b\u80fd\u529b\u3002\u53d6\u5411\u5f88\u6e05\u695a\uff1a\u5be7\u9858\u5148\u501f\u52a9\u5916\u90e8\u5de5\u5177\u62ff\u5230\u66f4\u53ef\u9760\u7684\u5c40\u90e8\u611f\u77e5\uff0c\u518d\u8ffd\u6c42\u628a\u80fd\u529b\u5167\u5316\uff0c\u6e1b\u5c11\u6bcf\u6b21\u63a8\u7406\u90fd\u4f9d\u8cf4\u5916\u639b\u6a21\u7d44\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">\u5c0d\u7814\u7a76\u5718\u968a\u6216\u505a\u591a\u6a21\u614b Agentic \u5de5\u4f5c\u6d41\u7684\u4eba\u4f86\u8aaa\uff0c\u9019\u500b\u9805\u76ee\u503c\u5f97\u7559\u610f\uff0c\u56e0\u70ba\u5b83\u540c\u6642\u653e\u51fa SpatialCLI code\u3001SpatialCLI-8B \u8207 SpatialCLI-Data\uff0c\u4e0d\u53ea\u662f\u6982\u5ff5\u5c55\u793a\u3002\u7406\u89e3\u5b83\u7684\u90e8\u7f72\u65b9\u5f0f\u4e5f\u4e0d\u96e3\uff1a\u4ee3\u78bc\u5eab\u8ca0\u8cac\u5de5\u5177\u8abf\u7528\u8207\u8a13\u7df4\u6d41\u7a0b\uff0cHugging Face \u4e0a\u7684\u6a21\u578b\u8207\u8cc7\u6599\u96c6\u5247\u5c0d\u61c9\u63a8\u7406\u3001\u5fae\u8abf\u548c\u91cd\u73fe\u5be6\u9a57\u7684\u6838\u5fc3\u6750\u6599\uff1b\u8981\u5b8c\u6574\u9a57\u8b49\u6548\u679c\uff0c\u901a\u5e38\u8981\u9023\u540c\u5916\u90e8\u7a7a\u9593\u5de5\u5177\u4e00\u4f75\u914d\u7f6e\u3002<\/p>\n\n\n\n<ul class=\"wp-block-list\">\n<li><strong>\u985e\u578b\u4e0a\u5c6c\u65bc\u6a21\u578b\u52a0\u6846\u67b6\u7684\u7814\u7a76\u9805\u76ee<\/strong>\uff0c\u76ee\u7684\u662f\u63d0\u5347\u591a\u6a21\u614b\u6a21\u578b\u5728\u7a7a\u9593\u63a8\u7406\u4e0a\u7684\u6e96\u78ba\u5ea6\u8207\u5de5\u5177\u4f7f\u7528\u80fd\u529b\u3002<\/li>\n<li>\u91cd\u9ede\u4e0d\u53ea\u5728\u300c\u53ef\u547c\u53eb\u5de5\u5177\u300d\uff0c\u800c\u662f\u9032\u4e00\u6b65\u628a\u5de5\u5177\u4f7f\u7528\u7d93\u9a57\u8f49\u5316\u6210\u6a21\u578b\u672c\u8eab\u7684\u80fd\u529b\u3002<\/li>\n<li>\u5df2\u516c\u958b\u8ad6\u6587\u3001SpatialCLI-8B \u8207 SpatialCLI-Data\uff0c\u65b9\u4fbf\u91cd\u73fe\u8207\u5ef6\u4f38\u8a13\u7df4\u3002<\/li>\n<li>\u9069\u5408\u8981\u8655\u7406\u5b9a\u4f4d\u3001\u5206\u5272\u3001\u6df1\u5ea6\u3001\u59ff\u614b\u7b49\u8996\u89ba\u4efb\u52d9\u7684\u4eba\u54e1\u53c3\u8003\u5176\u5de5\u4f5c\u6d41\u8a2d\u8a08\u3002<\/li>\n<\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">\u73fe\u6709\u8cc7\u8a0a\u672a\u898b README \u5b8c\u6574\u5217\u51fa\u91cf\u5316\u7d50\u679c\u7d30\u7bc0\uff0c\u4f46\u8a55\u6e2c\u7ae0\u7bc0\u8207 specialist models \u7ae0\u7bc0\u5df2\u9810\u7559\uff0c\u986f\u793a\u4f5c\u8005\u4e0d\u662f\u628a\u5b83\u5305\u88dd\u6210\u55ae\u4e00\u6a21\u578b\u5347\u7d1a\uff0c\u800c\u662f\u628a\u300c\u4f55\u6642\u8abf\u5de5\u5177\u3001\u5982\u4f55\u5b78\u6703\u4e0d\u7528\u5de5\u5177\u4e5f\u4fdd\u7559\u80fd\u529b\u300d\u7576\u6210\u6838\u5fc3\u554f\u984c\u3002\u9019\u7a2e\u505a\u6cd5\u7684\u4ee3\u50f9\u4e5f\u5f88\u660e\u986f\uff1a\u7cfb\u7d71\u6574\u5408\u8207\u8a13\u7df4\u93c8\u6703\u6bd4\u55ae\u7d14\u8dd1\u4e00\u500b VLM \u8907\u96dc\uff0c\u4e0d\u904e\u63db\u4f86\u7684\u662f\u66f4\u8cbc\u8fd1\u771f\u5be6\u7a7a\u9593\u4efb\u52d9\u7684\u63a8\u7406\u7a69\u5b9a\u6027\u3002<\/p>\n\n\n\n<p class=\"wp-block-paragraph\"><a href=\"https:\/\/github.com\/IANNXANG\/SpatialCLI\" rel=\"noopener noreferrer\"><strong>GitHub<\/strong><\/a><\/p>\n","protected":false},"excerpt":{"rendered":"<p>\u6a21\u578b\u5514\u4fc2\u7747\u5514\u660e\u5716\uff0c\u800c\u4fc2\u6210\u65e5\u5dee\u534a\u6b65\u5148\u7b54\u5f97\u6e96\u3002SpatialCLI \u60f3\u88dc\u4e0a\u7684\uff0c\u6b63\u6b63\u4fc2\u5b9a\u4f4d\u3001\u5206\u5272\u3001\u6df1\u5ea6\u8207\u59ff\u614b\u5224\u65b7\u5462\u985e\u5bb9\u6613\u5931\u624b\u7684\u7d30\u7bc0\u3002<\/p>\n","protected":false},"author":8,"featured_media":10595,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"ai_generated_summary":"","footnotes":""},"categories":[133,116,119,120,149],"tags":[],"class_list":["post-10596","post","type-post","status-publish","format-standard","hentry","category-133","category-agentic","category-119","category-120","category-149"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/10596","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/users\/8"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=10596"}],"version-history":[{"count":0,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/10596\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media\/10595"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=10596"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=10596"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=10596"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}