
{"id":4625,"date":"2025-02-17T05:08:17","date_gmt":"2025-02-16T21:08:17","guid":{"rendered":"https:\/\/infernews.com\/?p=4625"},"modified":"2025-02-17T09:03:23","modified_gmt":"2025-02-17T01:03:23","slug":"ktransformers-%e5%85%8b%e6%9c%8d%e8%b3%87%e6%ba%90%e9%99%90%e5%88%b6","status":"publish","type":"post","link":"https:\/\/infernews.com\/blog\/ktransformers-%e5%85%8b%e6%9c%8d%e8%b3%87%e6%ba%90%e9%99%90%e5%88%b6\/","title":{"rendered":"KTransformers \u514b\u670d\u8cc7\u6e90\u9650\u5236"},"content":{"rendered":"\n<p><a href=\"https:\/\/github.com\/kvcache-ai\/ktransformers\" target=\"_blank\" rel=\"noreferrer noopener\">KTransformers<\/a> \u900f\u904e\u5148\u9032\u7684\u6838\u5fc3\u512a\u5316\u6280\u8853\u4f86\u589e\u5f37 Transformers \u7684\u9ad4\u9a57\u3002KTransformers \u7279\u5225\u70ba\u786c\u4ef6\u8cc7\u6e90\u6709\u9650\u7684\u672c\u5730\u90e8\u7f72\uff0c\u4e26\u652f\u63f4\u7570\u69cb\u904b\u7b97\uff0c\u4f8b\u5982\u91cf\u5316\u6a21\u578b\u7684 GPU\/CPU \u5378\u8f09\uff0c\u4ee4\u5230\u4e0d\u540c\u7684\u67b6\u69cb\u8655\u7406\u5668\u53ef\u4ee5\u5354\u540c\u5de5\u4f5c\u3002KTransformers \u4ea6\u63d0\u4f9b\u4e86\u4e00\u500b YAML \u7bc4\u672c\u4f86\u547c\u53eb\u7279\u6b8a\u7684\u512a\u5316\u6307\u4ee4\u3002\u7ed3\u679c\u53ef\u4ee5\u4ee4 <strong>RTX 4090<\/strong> \u672c\u5730\u904b\u884c DeepSeek-R1\u3001V3 \u7684 671 B \u6ee1\u8840\u7248\uff0c\u4ee5 24 Gig VRam \u6700\u9ad8\u63a8\u7406\u901f\u5ea6\u4ecd\u9054\u5230 14 tokens\/s\u3002\u7576\u7136\uff0c\u4f60\u4ea6\u9700\u8981\u4fdd\u8a3c\u8db3\u5920 D Ram\u3002<\/p>\n\n\n\n<figure class=\"wp-block-audio\"><audio controls src=\"\/blog\/wp-content\/uploads\/2025\/02\/1739754133052169687-238390582575287.mp3\"><\/audio><\/figure>\n\n\n\n<figure class=\"wp-block-image size-full\"><a href=\"https:\/\/github.com\/kvcache-ai\/ktransformers\" target=\"_blank\"><img data-dominant-color=\"86868a\" data-has-transparency=\"false\" style=\"--dominant-color: #86868a;\" loading=\"lazy\" decoding=\"async\" width=\"1024\" height=\"562\" src=\"\/blog\/wp-content\/uploads\/2025\/02\/352761841-615dca9b-a08c-4183-bbd3-ad1362680faf.png\" alt=\"\" class=\"wp-image-4626 not-transparent\" srcset=\"https:\/\/infernews.com\/blog\/wp-content\/uploads\/2025\/02\/352761841-615dca9b-a08c-4183-bbd3-ad1362680faf.png 1024w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2025\/02\/352761841-615dca9b-a08c-4183-bbd3-ad1362680faf-300x165.png 300w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2025\/02\/352761841-615dca9b-a08c-4183-bbd3-ad1362680faf-768x422.png 768w, https:\/\/infernews.com\/blog\/wp-content\/uploads\/2025\/02\/352761841-615dca9b-a08c-4183-bbd3-ad1362680faf-640x351.png 640w\" sizes=\"auto, (max-width: 1024px) 100vw, 1024px\" \/><\/a><\/figure>\n","protected":false},"excerpt":{"rendered":"<p>KTransformers \u900f\u904e\u5148\u9032\u7684\u6838\u5fc3\u512a\u5316\u6280\u8853\u4f86\u589e\u5f37 Transformers \u7684\u9ad4\u9a57\u3002KTransformers \u7279\u5225\u70ba\u786c\u4ef6\u8cc7\u6e90\u6709\u9650\u7684\u672c\u5730\u90e8\u7f72\uff0c\u4e26\u652f\u63f4\u7570\u69cb\u904b\u7b97\uff0c\u4f8b\u5982\u91cf\u5316\u6a21\u578b\u7684 GPU\/CPU \u5378\u8f09\uff0c\u4ee4\u5230\u4e0d\u540c\u7684\u67b6\u69cb\u8655\u7406\u5668\u53ef\u4ee5\u5354\u540c\u5de5\u4f5c\u3002KTransformers \u4ea6\u63d0\u4f9b\u4e86\u4e00\u500b YAML \u7bc4\u672c\u4f86\u547c\u53eb\u7279\u6b8a\u7684\u512a\u5316\u6307\u4ee4\u3002\u7ed3\u679c\u53ef\u4ee5\u4ee4 RTX 4090 \u672c\u5730\u904b\u884c DeepSeek-R1\u3001V3 \u7684 671 B \u6ee1\u8840\u7248\uff0c\u4ee5 24 Gig VRam \u6700\u9ad8\u63a8\u7406\u901f\u5ea6\u4ecd\u9054\u5230 14 tokens\/s\u3002\u7576\u7136\uff0c\u4f60\u4ea6\u9700\u8981\u4fdd\u8a3c\u8db3\u5920 D Ram\u3002<\/p>\n","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"googlesitekit_rrm_CAowvqSiDA:productID":"","footnotes":""},"categories":[24,38],"tags":[],"class_list":["post-4625","post","type-post","status-publish","format-standard","hentry","category-aisoftware","category-38"],"_links":{"self":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/4625","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\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/comments?post=4625"}],"version-history":[{"count":0,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/posts\/4625\/revisions"}],"wp:attachment":[{"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/media?parent=4625"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/categories?post=4625"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/infernews.com\/blog\/wp-json\/wp\/v2\/tags?post=4625"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}