{"id":139,"date":"2024-07-23T13:22:46","date_gmt":"2024-07-23T05:22:46","guid":{"rendered":"https:\/\/www.onetts.com\/ai\/?p=139"},"modified":"2024-07-23T13:22:46","modified_gmt":"2024-07-23T05:22:46","slug":"yi-vl-6b","status":"publish","type":"post","link":"https:\/\/www.onetts.com\/ai\/models\/yi-vl-6b\/","title":{"rendered":"Yi-VL-6B"},"content":{"rendered":"<p data-spm-anchor-id=\"a2c6h.13066512.0.i3.243936afn78aKd\">Yi-VL-6B \u662f\u4e00\u4e2a\u5f00\u6e90\u7684\u591a\u6a21\u6001\u5927\u578b\u8bed\u8a00\u6a21\u578b\uff08LLM\uff09\uff0c\u7531\u96f6\u4e00\u4e07\u7269\u63a8\u51fa\u3002\u8be5\u6a21\u578b\u7ed3\u5408\u4e86\u89c6\u89c9\u548c\u8bed\u8a00\u5904\u7406\u80fd\u529b\uff0c\u80fd\u591f\u7406\u89e3\u548c\u751f\u6210\u6587\u672c\uff0c\u5e76\u5904\u7406\u4e0e\u56fe\u50cf\u76f8\u5173\u7684\u591a\u8f6e\u5bf9\u8bdd\u3002Yi-VL-6B \u5728\u591a\u6a21\u6001\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u8868\u73b0\u5353\u8d8a\uff0c\u7279\u522b\u662f\u5728 MMMU\uff08\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff09\u548c CMMMU\uff08\u4e2d\u6587\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff09\u4e2d\u6392\u540d\u7b2c\u4e00\u3002<\/p>\n<p><strong>\u4e3b\u8981\u7279\u70b9\uff1a<\/strong><\/p>\n<ul>\n<li><strong>\u591a\u8f6e\u6587\u672c-\u56fe\u50cf\u5bf9\u8bdd<\/strong>\uff1a\u80fd\u591f\u5904\u7406\u6587\u672c\u548c\u56fe\u50cf\u8f93\u5165\uff0c\u5e76\u751f\u6210\u6587\u672c\u8f93\u51fa\u3002<\/li>\n<li><strong>\u53cc\u8bed\u6587\u672c\u652f\u6301<\/strong>\uff1a\u652f\u6301\u82f1\u8bed\u548c\u4e2d\u6587\u7684\u5bf9\u8bdd\uff0c\u5305\u62ec\u56fe\u50cf\u4e2d\u7684\u6587\u5b57\u8bc6\u522b\u3002<\/li>\n<li><strong>\u5f3a\u5927\u7684\u56fe\u50cf\u7406\u89e3\u80fd\u529b<\/strong>\uff1a\u64c5\u957f\u5206\u6790\u89c6\u89c9\u4fe1\u606f\uff0c\u662f\u63d0\u53d6\u3001\u7ec4\u7ec7\u548c\u603b\u7ed3\u56fe\u50cf\u4fe1\u606f\u7684\u6709\u6548\u5de5\u5177\u3002<\/li>\n<li><strong>\u7ec6\u7c92\u5ea6\u56fe\u50cf\u5206\u8fa8\u7387<\/strong>\uff1a\u652f\u6301 448\u00d7448 \u7684\u9ad8\u5206\u8fa8\u7387\u56fe\u50cf\u7406\u89e3\u3002<\/li>\n<\/ul>\n<h2 id=\"-\">\u6a21\u578b\u8bc4\u6d4b<\/h2>\n<p>Yi-VL-6B \u5728\u591a\u4e2a\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u8868\u73b0\u51fa\u8272\uff0c\u7279\u522b\u662f\u5728\u4ee5\u4e0b\u4e24\u4e2a\u9886\u57df\uff1a<\/p>\n<ul>\n<li><strong>MMMU<\/strong>\uff1a\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff0c\u6d4b\u8bd5\u6a21\u578b\u5bf9\u591a\u5b66\u79d1\u591a\u6a21\u6001\u95ee\u9898\u7684\u7406\u89e3\u80fd\u529b\u3002<\/li>\n<li><strong>CMMMU<\/strong>\uff1a\u4e2d\u6587\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff0c\u4e13\u95e8\u6d4b\u8bd5\u6a21\u578b\u5bf9\u4e2d\u6587\u591a\u6a21\u6001\u95ee\u9898\u7684\u7406\u89e3\u80fd\u529b\u3002<\/li>\n<\/ul>\n<p>\u8fd9\u4e9b\u57fa\u51c6\u6d4b\u8bd5\u663e\u793a\uff0cYi-VL-6B \u5728\u5904\u7406\u590d\u6742\u591a\u6a21\u6001\u95ee\u9898\u65b9\u9762\u5177\u6709\u663e\u8457\u7684\u4f18\u52bf\u3002<\/p>\n<h2 id=\"-\">\u90e8\u7f72\u4f7f\u7528<\/h2>\n<h3 id=\"-\">\u786c\u4ef6\u8981\u6c42<\/h3>\n<ul>\n<li><strong>Yi-VL-6B<\/strong>\uff1a\u63a8\u8350\u4f7f\u7528 RTX 3090, RTX 4090, A10, A30 \u7b49 GPU\u3002<\/li>\n<\/ul>\n<h3 id=\"-\">\u90e8\u7f72\u6b65\u9aa4<\/h3>\n<ol>\n<li><strong>\u83b7\u53d6\u6a21\u578b<\/strong>\uff1a\u4ece ModelScope \u6216 Hugging Face \u4e0b\u8f7d Yi-VL-6B \u6a21\u578b\u3002<\/li>\n<li><strong>\u5b89\u88c5\u4f9d\u8d56<\/strong>\uff1a\u786e\u4fdd\u5b89\u88c5\u4e86\u5fc5\u8981\u7684\u5e93\u548c\u6846\u67b6\uff0c\u5982 PyTorch\u3002<\/li>\n<li><strong>\u52a0\u8f7d\u6a21\u578b<\/strong>\uff1a\u4f7f\u7528\u9002\u5f53\u7684\u52a0\u8f7d\u51fd\u6570\u52a0\u8f7d\u6a21\u578b\u3002<\/li>\n<li><strong>\u51c6\u5907\u8f93\u5165<\/strong>\uff1a\u5c06\u56fe\u50cf\u548c\u6587\u672c\u8f93\u5165\u8f6c\u6362\u4e3a\u6a21\u578b\u53ef\u63a5\u53d7\u7684\u683c\u5f0f\u3002<\/li>\n<li><strong>\u8fd0\u884c\u6a21\u578b<\/strong>\uff1a\u6267\u884c\u6a21\u578b\u63a8\u7406\uff0c\u83b7\u53d6\u8f93\u51fa\u7ed3\u679c\u3002<\/li>\n<\/ol>\n<h3 id=\"-\">\u793a\u4f8b\u4ee3\u7801<\/h3>\n<pre><code class=\"lang-python\"><span class=\"token keyword\">import<\/span> torch\r\n<span class=\"token keyword\">from<\/span> transformers <span class=\"token keyword\">import<\/span> AutoModel<span class=\"token punctuation\">,<\/span> AutoTokenizer\r\n\r\n<span class=\"token comment\" spellcheck=\"true\"># \u52a0\u8f7d\u6a21\u578b\u548c\u5206\u8bcd\u5668<\/span>\r\ntokenizer <span class=\"token operator\">=<\/span> AutoTokenizer<span class=\"token punctuation\">.<\/span>from_pretrained<span class=\"token punctuation\">(<\/span><span class=\"token string\">\"modelscope\/Yi-VL-6B\"<\/span><span class=\"token punctuation\">)<\/span>\r\nmodel <span class=\"token operator\">=<\/span> AutoModel<span class=\"token punctuation\">.<\/span>from_pretrained<span class=\"token punctuation\">(<\/span><span class=\"token string\">\"modelscope\/Yi-VL-6B\"<\/span><span class=\"token punctuation\">)<\/span>\r\n\r\n<span class=\"token comment\" spellcheck=\"true\"># \u51c6\u5907\u8f93\u5165<\/span>\r\ntext <span class=\"token operator\">=<\/span> <span class=\"token string\">\"What is in the image?\"<\/span>\r\nimage <span class=\"token operator\">=<\/span> <span class=\"token string\">\"path_to_your_image.jpg\"<\/span>\r\n\r\n<span class=\"token comment\" spellcheck=\"true\"># \u7f16\u7801\u6587\u672c\u548c\u56fe\u50cf<\/span>\r\ninputs <span class=\"token operator\">=<\/span> tokenizer<span class=\"token punctuation\">(<\/span>text<span class=\"token punctuation\">,<\/span> image<span class=\"token punctuation\">,<\/span> return_tensors<span class=\"token operator\">=<\/span><span class=\"token string\">\"pt\"<\/span><span class=\"token punctuation\">)<\/span>\r\n\r\n<span class=\"token comment\" spellcheck=\"true\"># \u8fd0\u884c\u6a21\u578b<\/span>\r\n<span class=\"token keyword\">with<\/span> torch<span class=\"token punctuation\">.<\/span>no_grad<span class=\"token punctuation\">(<\/span><span class=\"token punctuation\">)<\/span><span class=\"token punctuation\">:<\/span>\r\n    outputs <span class=\"token operator\">=<\/span> model<span class=\"token punctuation\">(<\/span><span class=\"token operator\">**<\/span>inputs<span class=\"token punctuation\">)<\/span>\r\n\r\n<span class=\"token comment\" spellcheck=\"true\"># \u83b7\u53d6\u7ed3\u679c<\/span>\r\nresult <span class=\"token operator\">=<\/span> outputs<span class=\"token punctuation\">.<\/span>logits\r\n<\/code><\/pre>\n<h2 id=\"-\">\u5e38\u89c1\u95ee\u9898<\/h2>\n<h3 id=\"q1-\">Q1: \u5982\u4f55\u5904\u7406\u4f4e\u5206\u8fa8\u7387\u56fe\u50cf\uff1f<\/h3>\n<p><strong>A1:<\/strong>\u00a0Yi-VL-6B \u8bad\u7ec3\u65f6\u4f7f\u7528\u7684\u662f 448\u00d7448 \u5206\u8fa8\u7387\u7684\u56fe\u50cf\u3002\u5728\u63a8\u7406\u65f6\uff0c\u6240\u6709\u8f93\u5165\u56fe\u50cf\u4f1a\u88ab\u8c03\u6574\u5230\u8fd9\u4e2a\u5206\u8fa8\u7387\u3002\u4f4e\u5206\u8fa8\u7387\u56fe\u50cf\u53ef\u80fd\u4f1a\u4e22\u5931\u4e00\u4e9b\u7ec6\u8282\uff0c\u800c\u9ad8\u4e8e 448\u00d7448 \u7684\u56fe\u50cf\u5219\u4e0d\u4f1a\u589e\u52a0\u989d\u5916\u7684\u77e5\u8bc6\u3002<\/p>\n<h3 id=\"q2-\">Q2: \u6a21\u578b\u662f\u5426\u652f\u6301\u591a\u56fe\u50cf\u8f93\u5165\uff1f<\/h3>\n<p><strong>A2:<\/strong>\u00a0\u76ee\u524d\uff0cYi-VL-6B \u53ea\u652f\u6301\u5355\u56fe\u50cf\u8f93\u5165\u3002\u672a\u6765\u53ef\u80fd\u4f1a\u589e\u52a0\u5bf9\u591a\u56fe\u50cf\u8f93\u5165\u7684\u652f\u6301\u3002<\/p>\n<h3 id=\"q3-\">Q3: \u5982\u4f55\u4f18\u5316\u5fae\u8c03\u6a21\u578b\u6027\u80fd\uff1f<\/h3>\n<p><strong>A3:<\/strong>\u00a0\u53ef\u4ee5\u901a\u8fc7\u8c03\u6574\u6a21\u578b\u53c2\u6570\u3001\u4f7f\u7528\u66f4\u5927\u7684\u6570\u636e\u96c6\u8fdb\u884c\u8bad\u7ec3\u6216\u4f7f\u7528\u66f4\u5f3a\u5927\u7684\u786c\u4ef6\u6765\u4f18\u5316\u6a21\u578b\u6027\u80fd\u3002<\/p>\n<h2 id=\"-\">\u76f8\u5173\u8d44\u6e90<\/h2>\n<h3 id=\"-\">\u6a21\u578b\u4e0b\u8f7d<\/h3>\n<ul>\n<li><strong>ModelScope<\/strong>:\u00a0<a target=\"_blank\" rel=\"noopener noreferrer nofollow\" href=\"https:\/\/modelscope.cn\/models\/01ai\/Yi-VL-6B\">Yi-VL-6B<\/a><\/li>\n<li><strong>Hugging Face<\/strong>:\u00a0<a target=\"_blank\" rel=\"noopener noreferrer nofollow\" href=\"https:\/\/huggingface.co\/01-ai\/Yi-VL-6B\">Yi-VL-6B<\/a><\/li>\n<\/ul>\n<h3 id=\"-\">\u8bb8\u53ef\u8bc1<\/h3>\n<ul>\n<li><strong>Yi Series Models Community License Agreement 2.1<\/strong>: \u9002\u7528\u4e8e\u5b66\u672f\u7814\u7a76\u548c\u5546\u4e1a\u7528\u9014\uff0c\u514d\u8d39\u4f7f\u7528\u9700\u7533\u8bf7\u3002<\/li>\n<\/ul>\n<p>\u901a\u8fc7\u8fd9\u4e9b\u8d44\u6e90\uff0c\u60a8\u53ef\u4ee5\u66f4\u6df1\u5165\u5730\u4e86\u89e3\u548c\u4f7f\u7528 Yi-VL-6B \u6a21\u578b\u3002<\/p>\n","protected":false},"excerpt":{"rendered":"<p>Yi-VL-6B \u662f\u4e00\u4e2a\u5f00\u6e90\u7684\u591a\u6a21\u6001\u5927\u578b\u8bed\u8a00\u6a21\u578b\uff08LLM\uff09\uff0c\u7531\u96f6\u4e00\u4e07\u7269\u63a8\u51fa\u3002\u8be5\u6a21\u578b\u7ed3\u5408\u4e86\u89c6\u89c9\u548c\u8bed\u8a00\u5904\u7406\u80fd\u529b\uff0c\u80fd\u591f\u7406\u89e3\u548c\u751f\u6210\u6587\u672c\uff0c\u5e76\u5904\u7406\u4e0e\u56fe\u50cf\u76f8\u5173\u7684\u591a\u8f6e\u5bf9\u8bdd\u3002Yi-VL-6B \u5728\u591a\u6a21\u6001\u57fa\u51c6\u6d4b\u8bd5\u4e2d\u8868\u73b0\u5353\u8d8a\uff0c\u7279\u522b\u662f\u5728 MMMU\uff08\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff09\u548c CMMMU\uff08\u4e2d\u6587\u591a\u6a21\u6001\u95ee\u9898\u7406\u89e3\uff09\u4e2d\u6392\u540d\u7b2c\u4e00\u3002 \u4e3b\u8981\u7279\u70b9\uff1a \u591a\u8f6e\u6587\u672c-\u56fe\u50cf\u5bf9\u8bdd\uff1a\u80fd\u591f\u5904\u7406\u6587\u672c\u548c\u56fe\u50cf\u8f93\u5165\uff0c\u5e76\u751f\u6210\u6587\u672c\u8f93\u51fa\u3002 \u53cc\u8bed\u6587\u672c\u652f\u6301\uff1a\u652f\u6301\u82f1\u8bed\u548c\u4e2d\u6587\u7684\u5bf9\u8bdd\uff0c\u5305\u62ec\u56fe\u50cf\u4e2d\u7684\u6587\u5b57\u8bc6\u522b\u3002 \u5f3a\u5927\u7684\u56fe\u50cf\u7406\u89e3\u80fd\u529b\uff1a\u64c5\u957f\u5206\u6790\u89c6\u89c9\u4fe1\u606f\uff0c\u662f\u63d0\u53d6\u3001\u7ec4\u7ec7\u548c\u603b\u7ed3\u56fe\u50cf\u4fe1\u606f\u7684\u6709\u6548\u5de5\u5177\u3002 \u7ec6\u7c92\u5ea6\u56fe\u50cf\u5206\u8fa8\u7387\uff1a\u652f<\/p>\n","protected":false},"author":1,"featured_media":140,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[12],"tags":[],"collection":[],"company":[22],"rank":[],"class_list":["post-139","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-multimodal","company-01-ai"],"_links":{"self":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/139","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/comments?post=139"}],"version-history":[{"count":1,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/139\/revisions"}],"predecessor-version":[{"id":141,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/posts\/139\/revisions\/141"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/media\/140"}],"wp:attachment":[{"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/media?parent=139"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/categories?post=139"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/tags?post=139"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/collection?post=139"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/company?post=139"},{"taxonomy":"rank","embeddable":true,"href":"https:\/\/www.onetts.com\/ai\/wp-json\/wp\/v2\/rank?post=139"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}