{"id":39115,"date":"2025-07-09T12:10:43","date_gmt":"2025-07-09T04:10:43","guid":{"rendered":"https:\/\/www.1ai.net\/?p=39115"},"modified":"2025-07-09T12:10:43","modified_gmt":"2025-07-09T04:10:43","slug":"%e9%a9%ac%e6%96%af%e5%85%8b%e5%ae%a3%e5%b8%83-grok-4-%e8%ae%a1%e5%88%92%e4%ba%8e-7-%e6%9c%88-10-%e6%97%a5%e7%9b%b4%e6%92%ad%e5%8f%91%e5%b8%83","status":"publish","type":"post","link":"https:\/\/www.1ai.net\/en\/39115.html","title":{"rendered":"Musk Announces Grok 4 Scheduled for July 10th Live Launch"},"content":{"rendered":"

July 9 News.Musk<\/a>Announce xA<\/a> will release its newest model on Wednesday at 8 p.m. PT (10 a.m. BST on July 10). Grok 4<\/a> series.<\/p>\n

\"ce139f2aj00sz466p0037d000u000asm\"<\/p>\n

And as previously revealed, the Grok 4 series will have Grok 4 and a programmed version, Grok-4 Code.<\/p>\n

The new version of the model will support multimodal input (text+image), 130k token context, structured output, mathematical reasoning and function calls. In particular, Grok 4 Code will be deeply integrated into the Cursor editor, with intelligent completion, debugging, and execution features, in an effort to create an AI assistant for \"active programming\".<\/p>\n

In terms of performance, according to X blogger @legit_api, Grok-4 achieved a base score of 35% in the evaluation dubbed the 'Humanity's Last Exam' (HLE), which was further boosted to 45% with inference turned on, significantly ahead of OpenAI's o3 and Google's Gemini series in that test.<\/p>\n

In addition to the HLE, Grok-4 performs well in other benchmarks. In the GPQA test, the model scored 87%-88%, while the SWE Bench review, which is geared towards code capability, scored the Grok-4 Code between 72%-75%.<\/p>\n

However, some netizens have questioned the authenticity of the scores, saying that the scores are \"too ideal\" and that they do not rule out the possibility of optimizing the testing strategy or even \"brushing the list\". Nonetheless, judging from the features and configurations that have been released so far, the Grok-4 series has been significantly upgraded compared to its predecessor.<\/p>\n

Also, Musk mentioned xAI's 3A masterpiece progress again yesterday:The first game could end by the end of next year.<\/strong>In February, Musk posted that he saw the possibility of making a game through Grok 3, so there's a chance that a 3A game could be made sometime next year.<\/p>","protected":false},"excerpt":{"rendered":"

On July 9, Musk announced that xAI will be releasing its newest model, the Grok 4 series, on Wednesday night at 8 p.m. Pacific Time (July 10 at 10 a.m. GMT). According to previous reports, the Grok 4 series will feature Grok 4 and a programmable version, Grok-4 Code. The new version of the model will support multimodal inputs (text+image), 130k token contexts, structured outputs, mathematical reasoning, and function calls, among other features. In particular, Grok 4 Code will be deeply integrated into the Cursor editor, with intelligent completion, debugging and execution functions, in an attempt to create an AI assistant for \"active programming\". In terms of performance, according to X blogger @<\/p>","protected":false},"author":1,"featured_media":0,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[146],"tags":[7064,356,355],"collection":[],"class_list":["post-39115","post","type-post","status-publish","format-standard","hentry","category-news","tag-grok-4","tag-xai","tag-355"],"acf":[],"_links":{"self":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/39115","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/users\/1"}],"replies":[{"embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/comments?post=39115"}],"version-history":[{"count":0,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/posts\/39115\/revisions"}],"wp:attachment":[{"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/media?parent=39115"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/categories?post=39115"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/tags?post=39115"},{"taxonomy":"collection","embeddable":true,"href":"https:\/\/www.1ai.net\/en\/wp-json\/wp\/v2\/collection?post=39115"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}