{"id":6106,"date":"2026-10-19T16:48:52","date_gmt":"2026-10-19T16:48:52","guid":{"rendered":"https:\/\/i10xnews.kinsta.cloud\/?p=6106"},"modified":"2026-10-19T17:15:42","modified_gmt":"2026-10-19T17:15:42","slug":"vgi-bench-video-generation-visual-reasoning","status":"publish","type":"post","link":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning","title":{"rendered":"Best Video Model Scores Just 51% on Visual Reasoning Tasks, VGI-Bench Finds"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">\u26a1 Quick Take<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>The finding:<\/strong> Even the strongest video generation model tested, Seedance 2.0, scores only 51.0% on VGI-Bench&#8217;s visual reasoning criteria.<\/li><li><strong>No real self-correction:<\/strong> During denoising, self-correction stays below 1%; wrong states more often morph into other wrong states (about 23\u201325%).<\/li><li><strong>What it means:<\/strong> Video models can &#8220;imagine&#8221; some step-by-step visual processes, but they are not reliable reasoners yet.<\/li><\/ul>\n\n\n<p class=\"wp-block-paragraph\">Some recent papers claim video generators show zero-shot visual reasoning by emitting intermediate frames. Researchers from UIUC, Tsinghua, Waterloo, MIT, UBC, Microsoft Research and others argue that existing tests are poorly calibrated. They introduce VGI-Bench: 27 tasks and 810 instances organized by domain and skill tags.<\/p>\n\n\n\n<figure class=\"wp-block-image size-large\"><img loading=\"lazy\" decoding=\"async\" width=\"1344\" height=\"768\" src=\"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-headline.webp\" alt=\"Stat card: 27 tasks, 810 instances, top score 51 percent for Seedance 2.0, self-correction under 1 percent during denoising.\" class=\"wp-image-6105\" srcset=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-headline.webp 1344w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-headline-300x171.webp 300w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-headline-1024x585.webp 1024w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-headline-768x439.webp 768w\" sizes=\"auto, (max-width: 1344px) 100vw, 1344px\" \/><figcaption class=\"wp-element-caption\">i10X chart based on data from He et al. (2026), arXiv:2608.19583, main evaluation and denoising analysis (CC BY 4.0).<\/figcaption><\/figure>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2c What the study shows<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Tasks require valid evolving processes, not only a plausible final frame, and use inputs aligned with video-model priors. A VLM-based evaluator (main setup with Gemini-3-Flash) reaches 73.2% agreement with humans; ablations with GPT-5-mini, Claude Haiku 4.5 and Qwen3.6-Plus score lower.<\/p>\n\n\n<ul class=\"wp-block-list\"><li><strong>Far from reliable:<\/strong> Current systems solve a subset of tasks but fail many others.<\/li><li><strong>Denoising mostly refines guesses:<\/strong> Later steps mainly polish early hypotheses; once a state is readable, stability rises to 90.6% in the second half of denoising, locking in early errors.<\/li><li><strong>Input sensitivity:<\/strong> Performance depends on input conditions; synthetic fine-tuning transfer has limits.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u2705 Why it matters: generation \u2260 reasoning<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Impressive videos can hide broken intermediate logic. VGI-Bench forces models to show the process, which is what robotics, planning and scientific visualization actually need.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83e\uddea Method in brief<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Two-level taxonomy of domains and skill tags; VLM evaluator validated against human annotations. Project page, data and code accompany the paper.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2753 What remains open<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Evaluator dependence:<\/strong> Scores shift with the judge model.<\/li><li><strong>Task design:<\/strong> Calibrated difficulty still reflects the authors&#8217; taxonomy.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\ud83e\udded Context<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Related multimodal evaluation work is in <a href=\"\/news\/category\/studies\/multimodal\">Multimodal<\/a> and <a href=\"\/news\/category\/studies\/benchmarks-evals\">Benchmarks &amp; Evals<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udee0\ufe0f What it means for video-AI teams<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Score processes, not finals:<\/strong> Require valid intermediate states.<\/li><li><strong>Watch early denoising:<\/strong> Errors lock in early; late steps rarely fix reasoning.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u270d\ufe0f About the analysis<\/h2>\n\n\n<p class=\"wp-block-paragraph\">This article is based on the full text of the preprint &#8220;VGI-Bench: Probing Visual Intelligence in Video Generation Models&#8221; (arXiv:2608.19583v3, submitted August 20, 2026). All numbers are taken from the paper&#8217;s text, tables and figures; i10X did not rerun the experiments. The chart was redrawn by i10X from main evaluation and denoising analysis of the paper, which is published under a CC BY 4.0 license.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2d i10x Perspective<\/h2>\n\n\n<p class=\"wp-block-paragraph\">51% for the best model is not a dunk on video AI so much as a useful reset. If we want models that reason by imagining, we need tests that punish plausible endings with impossible middles. VGI-Bench is that kind of test.<\/p>\n\n","protected":false},"excerpt":{"rendered":"<p>VGI-Bench&#8217;s 27 tasks and 810 instances show video generators can do some visual reasoning, but Seedance 2.0 tops out at 51%, with almost no self-correction during denoising.<\/p>\n","protected":false},"author":2,"featured_media":6256,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"paper_title":"VGI-Bench: Probing Visual Intelligence in Video Generation Models","paper_url":"https:\/\/arxiv.org\/abs\/2608.19583","pdf_url":"https:\/\/arxiv.org\/pdf\/2608.19583v3","doi":"","arxiv_id":"2608.19583","authors":"Xuan He, Cong Wei, Yuhao Cheng, Linrui Ma, Yuxuan Zhang, Zuojun Li, Yuhao Wen, Jize Jiang, Zeyi Liu, Yuren Hao, Songcheng Cai, Keming Wu et al. (23 authors)","institutions":"UIUC; Tsinghua; University of Waterloo; MIT; UBC; Vector Institute; Microsoft Research; others","paper_date":"2026-08-20","venue":"arXiv (cs.CV), version 3","peer_reviewed":false,"code_url":"","data_url":"","funding_coi":"Microsoft Research affiliations among authors.","figure_license":"Chart drawn by i10X from main evaluation and denoising analysis (CC BY 4.0)","paper_license":"CC BY 4.0","replication":"no","study_models":"Video generation models incl. Seedance 2.0; VLM judges Gemini-3-Flash, GPT-5-mini, Claude Haiku 4.5, Qwen3.6-Plus","study_sample":"27 tasks, 810 instances","footnotes":"","_yoast_wpseo_primary_category":""},"categories":[53,52],"tags":[],"companies":[39,37,38],"products":[51,49],"class_list":["post-6106","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-benchmarks-evals","category-studies","company-anthropic","company-google","company-open-ai","product-claude","product-gemini"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News<\/title>\n<meta name=\"description\" content=\"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News\" \/>\n<meta property=\"og:description\" content=\"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning\" \/>\n<meta property=\"og:site_name\" content=\"i10X News\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-19T16:48:52+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-10-19T17:15:42+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp\" \/>\n\t<meta property=\"og:image:width\" content=\"1344\" \/>\n\t<meta property=\"og:image:height\" content=\"768\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/webp\" \/>\n<meta name=\"author\" content=\"Christopher Ort\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@i10x_ai\" \/>\n<meta name=\"twitter:site\" content=\"@i10x_ai\" \/>\n<meta name=\"twitter:label1\" content=\"Written by\" \/>\n\t<meta name=\"twitter:data1\" content=\"Christopher Ort\" \/>\n\t<meta name=\"twitter:label2\" content=\"Est. reading time\" \/>\n\t<meta name=\"twitter:data2\" content=\"2 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\"},\"headline\":\"Best Video Model Scores Just 51% on Visual Reasoning Tasks, VGI-Bench Finds\",\"datePublished\":\"2026-10-19T16:48:52+00:00\",\"dateModified\":\"2026-10-19T17:15:42+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning\"},\"wordCount\":452,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/vgi-bench-video-generation-visual-reasoning-title.webp\",\"articleSection\":[\"Benchmarks &amp; Evals\",\"Studies\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning\",\"name\":\"VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/vgi-bench-video-generation-visual-reasoning-title.webp\",\"datePublished\":\"2026-10-19T16:48:52+00:00\",\"dateModified\":\"2026-10-19T17:15:42+00:00\",\"description\":\"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/vgi-bench-video-generation-visual-reasoning-title.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/vgi-bench-video-generation-visual-reasoning-title.webp\",\"width\":1344,\"height\":768,\"caption\":\"Illustration: i10X\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/vgi-bench-video-generation-visual-reasoning#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/news.i10x.ai\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Best Video Model Scores Just 51% on Visual Reasoning Tasks, VGI-Bench Finds\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\",\"url\":\"https:\\\/\\\/news.i10x.ai\\\/\",\"name\":\"i10X News\",\"description\":\"Stay updated with the latest news and announcements from i10X.\",\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/news.i10x.ai\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\",\"name\":\"i10X\",\"url\":\"https:\\\/\\\/news.i10x.ai\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/logo\\\/image\\\/\",\"url\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/i10x-logo-180.png\",\"contentUrl\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/i10x-logo-180.png\",\"width\":180,\"height\":180,\"caption\":\"i10X\"},\"image\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/logo\\\/image\\\/\"},\"sameAs\":[\"https:\\\/\\\/x.com\\\/i10x_ai\",\"https:\\\/\\\/www.linkedin.com\\\/company\\\/i10x-ai\",\"https:\\\/\\\/instagram.com\\\/i10x.ai\",\"https:\\\/\\\/www.youtube.com\\\/@i10X_ai_official\"]},{\"@type\":\"Person\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\",\"name\":\"Christopher Ort\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/christopher-ort-192.webp\",\"caption\":\"Christopher Ort\"},\"description\":\"Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU). He is the author of \u201cThe Style Penalty\u201d, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/author\\\/christopher-ort\",\"sameAs\":[\"https:\\\/\\\/i10x.ai\\\/blog\\\/author\\\/christopher-ort\"],\"worksFor\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#organization\"},\"alumniOf\":{\"@type\":\"EducationalOrganization\",\"name\":\"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU)\",\"url\":\"https:\\\/\\\/www.fau.eu\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News","description":"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.","robots":{"index":"index","follow":"follow","max-snippet":"max-snippet:-1","max-image-preview":"max-image-preview:large","max-video-preview":"max-video-preview:-1"},"canonical":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning","og_locale":"en_US","og_type":"article","og_title":"VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News","og_description":"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.","og_url":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning","og_site_name":"i10X News","article_published_time":"2026-10-19T16:48:52+00:00","article_modified_time":"2026-10-19T17:15:42+00:00","og_image":[{"width":1344,"height":768,"url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp","type":"image\/webp"}],"author":"Christopher Ort","twitter_card":"summary_large_image","twitter_creator":"@i10x_ai","twitter_site":"@i10x_ai","twitter_misc":{"Written by":"Christopher Ort","Est. reading time":"2 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#article","isPartOf":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning"},"author":{"name":"Christopher Ort","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a"},"headline":"Best Video Model Scores Just 51% on Visual Reasoning Tasks, VGI-Bench Finds","datePublished":"2026-10-19T16:48:52+00:00","dateModified":"2026-10-19T17:15:42+00:00","mainEntityOfPage":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning"},"wordCount":452,"commentCount":0,"publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"image":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp","articleSection":["Benchmarks &amp; Evals","Studies"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#respond"]}]},{"@type":"WebPage","@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning","url":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning","name":"VGI-Bench: Video Generators Still Unreliable at Visual Reasoning - i10X News","isPartOf":{"@id":"https:\/\/news.i10x.ai\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#primaryimage"},"image":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp","datePublished":"2026-10-19T16:48:52+00:00","dateModified":"2026-10-19T17:15:42+00:00","description":"VGI-Bench: 27 tasks, 810 instances. Strongest video model Seedance 2.0 scores 51%; self-correction during denoising stays under 1%.","breadcrumb":{"@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#primaryimage","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/vgi-bench-video-generation-visual-reasoning-title.webp","width":1344,"height":768,"caption":"Illustration: i10X"},{"@type":"BreadcrumbList","@id":"https:\/\/i10x.ai\/news\/vgi-bench-video-generation-visual-reasoning#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/news.i10x.ai\/"},{"@type":"ListItem","position":2,"name":"Best Video Model Scores Just 51% on Visual Reasoning Tasks, VGI-Bench Finds"}]},{"@type":"WebSite","@id":"https:\/\/news.i10x.ai\/#website","url":"https:\/\/news.i10x.ai\/","name":"i10X News","description":"Stay updated with the latest news and announcements from i10X.","publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/news.i10x.ai\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/news.i10x.ai\/#organization","name":"i10X","url":"https:\/\/news.i10x.ai\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/news.i10x.ai\/#\/schema\/logo\/image\/","url":"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/i10x-logo-180.png","contentUrl":"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/10\/i10x-logo-180.png","width":180,"height":180,"caption":"i10X"},"image":{"@id":"https:\/\/news.i10x.ai\/#\/schema\/logo\/image\/"},"sameAs":["https:\/\/x.com\/i10x_ai","https:\/\/www.linkedin.com\/company\/i10x-ai","https:\/\/instagram.com\/i10x.ai","https:\/\/www.youtube.com\/@i10X_ai_official"]},{"@type":"Person","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a","name":"Christopher Ort","image":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/christopher-ort-192.webp","caption":"Christopher Ort"},"description":"Christopher Ort handles SEO, AI agent workflows and PR & partnerships at i10x.ai, the all-in-one AI workspace built by i10X Global Pte. Ltd. in Singapore. He is a graduate of Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU). He is the author of \u201cThe Style Penalty\u201d, a study published on SSRN that analyses 1,576 evaluation decisions by four frontier AI models to show how AI-written CVs fare in automated hiring screens; the research has been covered by heise online and t3n.","url":"https:\/\/i10x.ai\/news\/author\/christopher-ort","sameAs":["https:\/\/i10x.ai\/blog\/author\/christopher-ort"],"worksFor":{"@id":"https:\/\/i10x.ai\/news\/#organization"},"alumniOf":{"@type":"EducationalOrganization","name":"Friedrich-Alexander-Universit\u00e4t Erlangen-N\u00fcrnberg (FAU)","url":"https:\/\/www.fau.eu"}}]}},"_links":{"self":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6106","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/users\/2"}],"replies":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/comments?post=6106"}],"version-history":[{"count":2,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6106\/revisions"}],"predecessor-version":[{"id":6257,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6106\/revisions\/6257"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media\/6256"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media?parent=6106"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/categories?post=6106"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/tags?post=6106"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/companies?post=6106"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/products?post=6106"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}