{"id":6108,"date":"2026-10-19T16:48:53","date_gmt":"2026-10-19T16:48:53","guid":{"rendered":"https:\/\/i10xnews.kinsta.cloud\/?p=6108"},"modified":"2026-10-19T17:15:57","modified_gmt":"2026-10-19T17:15:57","slug":"fde-gym-ai-delivery-last-mile-benchmark","status":"publish","type":"post","link":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark","title":{"rendered":"AI Agents That Learn the Facts Still Deliver Only 60\u201370% of the Business Value"},"content":{"rendered":"\n<h2 class=\"wp-block-heading\">\u26a1 Quick Take<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>The finding:<\/strong> On 95 real-world delivery tasks, Claude Opus 5.5 leads with a mean score of 0.56 (95% CI 0.50\u20130.62). GPT-6 Astra and GPT-6.1 Sol trail at 0.38 and 0.37.<\/li><li><strong>Facts \u2260 delivery:<\/strong> Leading models obtain 72\u201385% of key stakeholder facts yet deliver only 0.60\u20130.70 of the reference business improvement.<\/li><li><strong>What it means:<\/strong> Agent benches that hand over a full ticket miss the hard part: discovering constraints and turning them into working systems.<\/li><\/ul>\n\n\n<p class=\"wp-block-paragraph\">Most agent benchmarks give a complete issue report or ticket. Real delivery work starts with a vague sponsor request. FDE-Gym, from DeepLethe and collaborators across universities and industry, adapts 20 real delivery projects from 18 sectors into customer environments, then creates variants for 95 grader-validated tasks.<\/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\/fde-gym-ai-delivery-last-mile-benchmark-l3-scores.webp\" alt=\"Bar chart of L3 FDE-Gym scores: Claude Opus 5.5 at 0.56, Sonnet 5.5 at 0.49, GPT-6 Astra 0.38, GPT-6.1 Sol 0.37.\" class=\"wp-image-6107\" srcset=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-l3-scores.webp 1344w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-l3-scores-300x171.webp 300w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-l3-scores-1024x585.webp 1024w, https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-l3-scores-768x439.webp 768w\" sizes=\"auto, (max-width: 1344px) 100vw, 1344px\" \/><figcaption class=\"wp-element-caption\">i10X chart based on data from Yang et al. (2026), arXiv:2610.22350, L3 evaluation (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\">Instructions come in levels L0\u2013L3 with progressively less information. Scores are normalized so 0 is do-nothing and 1 is the reference solution, with stakeholder contact cost deducted. Human baselines are still being collected.<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table><thead><tr><th>Model at L3 (all 95 tasks)<\/th><th>Mean score<\/th><\/tr><\/thead><tbody><tr><td>Claude Opus 5.5<\/td><td>0.56 [0.50, 0.62]<\/td><\/tr><tr><td>Claude Sonnet 5.5<\/td><td>0.49<\/td><\/tr><tr><td>GPT-6 Astra<\/td><td>0.38<\/td><\/tr><tr><td>GPT-6.1 Sol<\/td><td>0.37<\/td><\/tr><\/tbody><\/table><figcaption class=\"wp-element-caption\">Mean FDE-Gym scores with stakeholder contact cost deducted.<\/figcaption><\/figure>\n\n\n<ul class=\"wp-block-list\"><li><strong>Constraints matter most:<\/strong> Withholding constraints drops the three leading models by 0.28 on the same tasks (95% CI 0.17\u20130.39).<\/li><li><strong>Questions help, imperfectly:<\/strong> Top models ask 10.0\u201312.3 questions per task. Even with every key fact, uplift averages 0.66 and hits \u22650.8 in only 37% of those cases.<\/li><li><strong>Weak models make things worse:<\/strong> The weakest models score below the do-nothing baseline on 60\u201383% of tasks.<\/li><li><strong>Silent non-askers:<\/strong> Mistral Small 4 asks 0.7 questions per task and obtains only 6% of key facts.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\u2705 Why it matters: the last mile is the product<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Enterprises do not buy agents that pass unit tests on fully specified tickets. They buy delivery under incomplete briefs. FDE-Gym measures that gap.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83e\uddea Method in brief<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Experts adapt real projects, alter facts for variants, and validate graders against scoring anchors. Bootstrapped 95% CIs over tasks for L3.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u2753 What remains open<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Human baseline pending:<\/strong> Expert performance under L3 is still being measured.<\/li><li><strong>Contact-cost modeling:<\/strong> Stakeholder questions are costed; real org politics are richer.<\/li><\/ul>\n\n\n<h2 class=\"wp-block-heading\">\ud83e\udded Context<\/h2>\n\n\n<p class=\"wp-block-paragraph\">Complements coding and GUI agent benches by scoring business uplift. See <a href=\"\/news\/category\/studies\/benchmarks-evals\">Benchmarks &amp; Evals<\/a> and <a href=\"\/news\/category\/studies\/agent-research\">Agents<\/a>.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udee0\ufe0f What it means for AI delivery teams<\/h2>\n\n<ul class=\"wp-block-list\"><li><strong>Train constraint discovery:<\/strong> Missing constraints hurt more than missing niceties.<\/li><li><strong>Measure uplift, not tickets closed:<\/strong> A finished system can still score below do-nothing.<\/li><li><strong>Force clarifying questions:<\/strong> Models that rarely ask fail early.<\/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;FDE-Gym: The Last Mile of AI Delivery&#8221; (arXiv:2610.22350v1, submitted October 15, 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 L3 evaluation 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\">The most damning number is not the leaderboard: models that learn almost all the facts still deliver only about two-thirds of the reference improvement. Knowledge gathering is necessary and nowhere near sufficient. That is the last mile of AI delivery.<\/p>\n\n","protected":false},"excerpt":{"rendered":"<p>FDE-Gym&#8217;s 95 real delivery tasks: Claude Opus 5.5 leads at 0.56, but even models that gather 72\u201385% of key facts deliver just 0.60\u20130.70 of the reference uplift.<\/p>\n","protected":false},"author":2,"featured_media":6258,"comment_status":"open","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"paper_title":"FDE-Gym: The Last Mile of AI Delivery","paper_url":"https:\/\/arxiv.org\/abs\/2610.22350","pdf_url":"https:\/\/arxiv.org\/pdf\/2610.22350v1","doi":"","arxiv_id":"2610.22350","authors":"Dongxu Yang, Hongda Yan, Renxin Wang, Baoli Zhang, Haojia Zuo, Ying Zhao, Youmin Chen, Xiaojian Liao, Congming Gao, Tianyang Jiang, Yixiao Chen, Qing Wang et al. (26 authors)","institutions":"DeepLethe; Tsinghua; Shanghai Jiao Tong; Beihang; Xiamen; Nanjing; many others incl. Meta, Microsoft Research Asia, Huawei","paper_date":"2026-10-15","venue":"arXiv (cs.AI), version 1","peer_reviewed":false,"code_url":"","data_url":"","funding_coi":"Industry and academic affiliations; see paper.","figure_license":"Chart drawn by i10X from L3 evaluation (CC BY 4.0)","paper_license":"CC BY 4.0","replication":"no","study_models":"Claude Opus\/Sonnet 5.5, GPT-6 Astra, GPT-6.1 Sol, Mistral Small 4 and others","study_sample":"95 grader-validated tasks from 20 real delivery projects across 18 sectors","footnotes":"","_yoast_wpseo_primary_category":""},"categories":[53,52],"tags":[],"companies":[39,45,38],"products":[51],"class_list":["post-6108","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-benchmarks-evals","category-studies","company-anthropic","company-mistral","company-open-ai","product-claude"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News<\/title>\n<meta name=\"description\" content=\"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.\" \/>\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\/fde-gym-ai-delivery-last-mile-benchmark\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News\" \/>\n<meta property=\"og:description\" content=\"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark\" \/>\n<meta property=\"og:site_name\" content=\"i10X News\" \/>\n<meta property=\"article:published_time\" content=\"2026-10-19T16:48:53+00:00\" \/>\n<meta property=\"article:modified_time\" content=\"2026-10-19T17:15:57+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-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\\\/fde-gym-ai-delivery-last-mile-benchmark#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\"},\"headline\":\"AI Agents That Learn the Facts Still Deliver Only 60\u201370% of the Business Value\",\"datePublished\":\"2026-10-19T16:48:53+00:00\",\"dateModified\":\"2026-10-19T17:15:57+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark\"},\"wordCount\":508,\"commentCount\":0,\"publisher\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/fde-gym-ai-delivery-last-mile-benchmark-title.webp\",\"articleSection\":[\"Benchmarks &amp; Evals\",\"Studies\"],\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"CommentAction\",\"name\":\"Comment\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#respond\"]}]},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark\",\"name\":\"FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/news.i10x.ai\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/fde-gym-ai-delivery-last-mile-benchmark-title.webp\",\"datePublished\":\"2026-10-19T16:48:53+00:00\",\"dateModified\":\"2026-10-19T17:15:57+00:00\",\"description\":\"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/fde-gym-ai-delivery-last-mile-benchmark-title.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/10\\\/fde-gym-ai-delivery-last-mile-benchmark-title.webp\",\"width\":1344,\"height\":768,\"caption\":\"Illustration: i10X\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/fde-gym-ai-delivery-last-mile-benchmark#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/news.i10x.ai\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"AI Agents That Learn the Facts Still Deliver Only 60\u201370% of the Business Value\"}]},{\"@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":"FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News","description":"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.","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\/fde-gym-ai-delivery-last-mile-benchmark","og_locale":"en_US","og_type":"article","og_title":"FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News","og_description":"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.","og_url":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark","og_site_name":"i10X News","article_published_time":"2026-10-19T16:48:53+00:00","article_modified_time":"2026-10-19T17:15:57+00:00","og_image":[{"width":1344,"height":768,"url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-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\/fde-gym-ai-delivery-last-mile-benchmark#article","isPartOf":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark"},"author":{"name":"Christopher Ort","@id":"https:\/\/news.i10x.ai\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a"},"headline":"AI Agents That Learn the Facts Still Deliver Only 60\u201370% of the Business Value","datePublished":"2026-10-19T16:48:53+00:00","dateModified":"2026-10-19T17:15:57+00:00","mainEntityOfPage":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark"},"wordCount":508,"commentCount":0,"publisher":{"@id":"https:\/\/news.i10x.ai\/#organization"},"image":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-title.webp","articleSection":["Benchmarks &amp; Evals","Studies"],"inLanguage":"en-US","potentialAction":[{"@type":"CommentAction","name":"Comment","target":["https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#respond"]}]},{"@type":"WebPage","@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark","url":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark","name":"FDE-Gym: The Last Mile of AI Delivery Is Still Hard - i10X News","isPartOf":{"@id":"https:\/\/news.i10x.ai\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage"},"image":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-title.webp","datePublished":"2026-10-19T16:48:53+00:00","dateModified":"2026-10-19T17:15:57+00:00","description":"FDE-Gym: 95 expert-adapted delivery tasks. Claude Opus 5.5 leads at 0.56; knowing the facts is not enough to deliver the value.","breadcrumb":{"@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#primaryimage","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-title.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/fde-gym-ai-delivery-last-mile-benchmark-title.webp","width":1344,"height":768,"caption":"Illustration: i10X"},{"@type":"BreadcrumbList","@id":"https:\/\/i10x.ai\/news\/fde-gym-ai-delivery-last-mile-benchmark#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/news.i10x.ai\/"},{"@type":"ListItem","position":2,"name":"AI Agents That Learn the Facts Still Deliver Only 60\u201370% of the Business Value"}]},{"@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\/6108","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=6108"}],"version-history":[{"count":2,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6108\/revisions"}],"predecessor-version":[{"id":6259,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/6108\/revisions\/6259"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media\/6258"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media?parent=6108"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/categories?post=6108"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/tags?post=6108"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/companies?post=6108"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/products?post=6108"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}