{"id":8929,"date":"2026-11-14T07:28:50","date_gmt":"2026-11-14T07:28:50","guid":{"rendered":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims"},"modified":"2026-11-14T07:28:50","modified_gmt":"2026-11-14T07:28:50","slug":"reason-machines-coding-harness-cost-claims","status":"publish","type":"post","link":"https:\/\/i10x.ai\/news\/reason-machines-coding-harness-cost-claims","title":{"rendered":"Reason Coding Harness Promises 20x Lower Costs. Its Own Benchmarks Show 22%"},"content":{"rendered":"\n<p class=\"wp-block-paragraph\">Every team running coding agents knows the feeling: the agent fixed the bug, and the invoice fixed your mood. Reason Machines, a Y Combinator startup, is selling the cure. On November 13, 2026, cofounder Adi Singh <a href=\"https:\/\/x.com\/xadisingh\/status\/2121265353867284498\" target=\"_blank\" rel=\"noreferrer noopener\">introduced Reason on X<\/a> as &#8220;the world&#8217;s most cost-efficient harness,&#8221; promising &#8220;frontier coding performance at up to 20x lower cost.&#8221; The post took off. The company&#8217;s own published benchmarks, however, tell a more modest and more interesting story: a 22% lower cost per solved task than OpenAI&#8217;s Codex on the same model.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">\u26a1 Quick Take<\/h2>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>What launched:<\/strong> Reason, a coding agent and &#8220;harness&#8221; (the software around an AI model that manages prompts, tools and context) from Reason Machines, promoted on X on November 13, 2026.<\/li><li><strong>The headline claim:<\/strong> &#8220;Up to 20x lower cost&#8221; and better results than top harnesses on Terminal-Bench and DeepSWE, per the launch post.<\/li><li><strong>The published numbers:<\/strong> On GPT-6 Astra, Reason scored 65.1 against Codex&#8217;s 62.5 on the Coding Agent Index v1.5 at 22% lower cost per solved task, in the company&#8217;s own reproduction, which Artificial Analysis has not reviewed or endorsed.<\/li><li><strong>How it saves money:<\/strong> A small, stable prompt that maximizes prompt caching, and a &#8220;codemode&#8221; tool that runs several tool calls in one step.<\/li><li><strong>Price:<\/strong> Pro costs $50 per month and Max $200 per month, with usage billed in credits; users can also connect their own ChatGPT, Claude, Copilot or Grok subscriptions and API keys.<\/li><\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">In plain English<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">AI coding assistants charge for every chunk of text they read and write, and a lot of that text is the same instructions sent over and over. Reason is a new tool that wraps around existing AI models and tries to reuse that repeated text so it is billed at a discount. The company says this makes coding work much cheaper. Its own detailed tests show savings of roughly a fifth compared with OpenAI&#8217;s tool, which is useful but far from the &#8220;20 times cheaper&#8221; in its marketing post.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What happened<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Reason Machines was founded by Adi and Sven, who studied EECS at UC Berkeley and, in their words, &#8220;dropped out to build what a self-driving company runs on,&#8221; according to the company&#8217;s <a href=\"https:\/\/reasonmachines.com\/blog\/launch\" target=\"_blank\" rel=\"noreferrer noopener\">launch blog post<\/a> dated November 9. Singh&#8217;s X bio lists the company as part of YC F26. The blog post claims &#8220;20,000 users, 14,000 X followers, and 3M+ impressions&#8221; in the four weeks since launch, and says the company is raising its seed round.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The November 13 video post pushed the product into the trending list. In it, Reason patches a security bug, merges the fix and repairs a customer&#8217;s broken export straight from Slack. TestingCatalog summarized that Reason &#8220;solved more tasks than Codex on the same model while spending 22% less&#8221; and that users can give it standing jobs, such as reviewing every new pull request.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">The numbers, side by side<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Here is where the claims diverge. The launch post on X says &#8220;up to 20x.&#8221; The November 9 blog post says &#8220;up to 30% lower cost per task than Codex, Cursor or Devin.&#8221; The detailed <a href=\"https:\/\/reasonmachines.com\/blog\/introducing-reason-agent-1-1\" target=\"_blank\" rel=\"noreferrer noopener\">Reason 1.1 benchmark write-up<\/a> from October 10 reports:<\/p>\n\n\n\n<figure class=\"wp-block-table\"><table class=\"has-fixed-layout\"><thead><tr><th>Source<\/th><th>Date<\/th><th>Claim<\/th><\/tr><\/thead><tbody><tr><td>Launch post on X (Adi Singh)<\/td><td>Nov 13, 2026<\/td><td>&#8220;Up to 20x lower cost&#8221;; outperforms top harnesses on Terminal-Bench and DeepSWE<\/td><\/tr><tr><td>Launch blog post<\/td><td>Nov 9, 2026<\/td><td>Up to 30% lower cost per task than Codex, Cursor or Devin<\/td><\/tr><tr><td>Reason 1.1 write-up (GPT-6 Astra)<\/td><td>Oct 10, 2026<\/td><td>Coding Agent Index v1.5: Reason 65.1, Codex 62.5, Pi 61.4<\/td><\/tr><tr><td>Reason 1.1 write-up (cost)<\/td><td>Oct 10, 2026<\/td><td>Cost per solved task 22% below Codex and 10% below Pi (own reproduction, not endorsed by Artificial Analysis)<\/td><\/tr><\/tbody><\/table><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">The 1.1 write-up also says Reason &#8220;costs 15\u201323% less per attempt than Codex on every benchmark&#8221; and uses 23% fewer input tokens per attempt (1.92 million against 2.48 million) and 15% fewer output tokens on GPT-6 Astra. Crucially, all results are on Artificial Analysis&#8217;s Coding Agent Index v1.5 (DeepSWE v1.1, Terminal-Bench 4.0 and SWE-Atlas QnA). Artificial Analysis moved to version 2.0 on October 22, so the figures are not directly comparable with its current leaderboard. In the materials i10x reviewed, Reason did not publish a same-model comparison showing a 20x saving. A gap that large likely depends on comparing a cheap model inside Reason with an expensive frontier setup, but the company has not said so explicitly.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">How it works<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The core idea fits in one sentence from the company: &#8220;The model sets the pass rate; the harness sets the bill.&#8221; Reason says the same model can cost up to five times more in one harness than in another. Two design choices drive its savings:<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>Cache-friendly prompts:<\/strong> &#8220;The prompt and tools never change between turns and each request only appends to the last, so every turn is fully cache-eligible up to its new content.&#8221; As AI researcher Rohan Paul explained on X, providers bill repeated prompt prefixes at a steep discount, which compounds over long agent sessions.<\/li><li><strong>Codemode:<\/strong> A tool that runs JavaScript calling the other tools, so &#8220;a search, five file reads and a test run can take one model call.&#8221; Fewer round trips mean fewer tokens.<\/li><\/ul>\n\n\n\n<p class=\"wp-block-paragraph\">Everything else (subagents, cloud computers, preview URLs, pull requests with screenshots) sits behind &#8220;deferred tools&#8221; the model finds on demand, keeping the base prompt small. Reason also publishes its benchmark runner on <a href=\"https:\/\/github.com\/reason-machines\/benchmark\" target=\"_blank\" rel=\"noreferrer noopener\">GitHub<\/a>, so anyone can reproduce a run with the Harbor evaluation framework:<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The repository&#8217;s quick check, as a script we call <code>run-deepswe.sh<\/code>:<\/p>\n\n\n\n<figure class=\"nx-code\" data-lang=\"bash\"><div class=\"nx-code__bar\"><span class=\"nx-code__label\">run-deepswe.sh \u00b7 Bash<\/span><button type=\"button\" class=\"nx-code__copy\" aria-label=\"Copy code\" onclick=\"(function(b){var t=b.closest('.nx-code').querySelector('code').innerText;navigator.clipboard.writeText(t).then(function(){var o=b.textContent;b.textContent='Copied';setTimeout(function(){b.textContent=o},1500)})})(this)\">Copy<\/button><\/div><pre class=\"nx-code__pre nx-code__pre--nums\" style=\"--nx-ln-start:1\" tabindex=\"0\"><code class=\"language-bash\"><span class=\"nx-code__line\">uv tool install <span class=\"nx-tok-str\">&quot;harbor[e2b,modal]==0.22.0&quot;<\/span> --with <span class=\"nx-tok-str\">&quot;reason-machines-harbor&gt;=0.3.44&quot;<\/span><\/span>\n<span class=\"nx-code__line\"><span class=\"nx-tok-kw\">export<\/span> REASON_API_KEY=&lt;YOUR-REASON-KEY&gt;<\/span>\n<span class=\"nx-code__line\">&nbsp;<\/span>\n<span class=\"nx-code__line\"><span class=\"nx-tok-cm\"># One DeepSWE task on local Docker, to check your setup.<\/span><\/span>\n<span class=\"nx-code__line\">harbor run --dataset datacurve\/deep-swe-<span class=\"nx-tok-num\">1<\/span>-<span class=\"nx-tok-num\">1<\/span> --n-tasks <span class=\"nx-tok-num\">1<\/span> \\<\/span>\n<span class=\"nx-code__line\">   --agent reasonmachines_harbor:ReasonAgent \\<\/span>\n<span class=\"nx-code__line\">   --model openai\/gpt-<span class=\"nx-tok-num\">6<\/span>-luna \\<\/span>\n<span class=\"nx-code__line\">   --ak reasoning_effort=xhigh \\<\/span>\n<span class=\"nx-code__line\">   --allow-agent-host api.reasonmachines.com \\<\/span>\n<span class=\"nx-code__line\">   --allow-agent-host relay.reasonmachines.com<\/span><\/code><\/pre><\/figure>\n\n\n\n<p class=\"wp-block-paragraph\">Sessions run from a web app, CLI, macOS or Linux desktop app, Slack or an HTTP API, on cloud Linux, a cloud Mac, the user&#8217;s own Mac or an SSH box. Users can bring API keys, open-weight models and subscriptions from Grok, Copilot or Kimi with automatic failover, according to the launch post.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Pros and cons<\/h2>\n\n\n\n<h3 class=\"wp-block-heading\">For users and solo developers<\/h3>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>Pro:<\/strong> If the caching savings hold, a flat $50 per month with bring-your-own subscriptions can stretch existing AI plans further.<\/li><li><strong>Con:<\/strong> Another subscription on top of the ones you already pay for, and savings depend heavily on which model you plug in.<\/li><\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">For developers and engineering teams<\/h3>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>Pro:<\/strong> Reproducible benchmark code, Slack-native standing jobs (like reviewing every pull request) and support for many models instead of a single vendor.<\/li><li><strong>Con:<\/strong> The headline numbers are self-reported, and the benchmark version they use has already been replaced by Artificial Analysis.<\/li><\/ul>\n\n\n\n<h3 class=\"wp-block-heading\">For companies<\/h3>\n\n\n\n<ul class=\"wp-block-list\"><li><strong>Pro:<\/strong> Agent spend is becoming a real budget line. Even a 20% to 30% cut on large agent workloads adds up.<\/li><li><strong>Con:<\/strong> A seed-stage vendor in the middle of code, credentials and cloud machines is a security and continuity risk that procurement teams will want to examine carefully.<\/li><\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">What X is saying<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The launch post itself:<\/p>\n\n\n\n<blockquote class=\"twitter-tweet i10x-x-embed\"><p lang=\"en\" dir=\"ltr\">Frontier coding performance at up to 20x lower cost.<br><br>Introducing the world\u2019s most cost-efficient harness: Reason<br><br>Stop burning thousands on runaway agent bills. <br><br>Reason outperforms top harnesses on Terminal-Bench and DeepSWE while dropping cost per task to a fraction of the frontier baseline.<br><br>Comment \u201cReason\u201d and we\u2019ll send you early access + free compute credits.<\/p>&mdash; Adi Singh (@xadisingh) <a href=\"https:\/\/x.com\/xadisingh\/status\/2121265353867284498\" target=\"_blank\" rel=\"noreferrer noopener\">November 13, 2026 \u00b7 View on X<\/a><\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">TestingCatalog focused on the measured numbers:<\/p>\n\n\n\n<blockquote class=\"twitter-tweet i10x-x-embed\"><p lang=\"en\" dir=\"ltr\"> TestingCatalog (@testingcatalog)<\/p>&mdash; \ud83d\udea8 AI News  <a href=\"https:\/\/x.com\/testingcatalog\/status\/2121338593675489338\" target=\"_blank\" rel=\"noreferrer noopener\">Reason Machines has launched Reason, its AI coding agent designed to maximize work completed per dollar.\\n\\nDuring testing, Reason solved more tasks than Codex on the same model while spending 22% less.\\n\\nUsers can give it standing jobs to run autonomously, such as reviewing every new pull request, and run it using their own ChatGPT, Claude, GitHub Copilot, or Grok membership.|November 13, 2026 \u00b7 View on X<\/a><\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">Rohan Paul explained the caching trick:<\/p>\n\n\n\n<blockquote class=\"twitter-tweet i10x-x-embed\"><p lang=\"en\" dir=\"ltr\">Reason Machines just launched Reason, a coding harness that wraps a single model like GPT-6 Astra and solved tasks 22% cheaper than OpenAI\u2019s Codex.<br><br>How: A traditional coding agent resends its context to the model on every step, so Reason trims that load: <br><br>Reason holds the prompt prefix identical between turns, by putting its fixed instructions and small tool list at the top of every prompt and only adding new code and results at the end, never rewriting or reshuffling the earlier text.<br><br>Since every turn starts with the exact same text, OpenAI recognizes it from the previous turn and bills those repeated tokens 90% cheaper, at $1 instead of $10 per million.<\/p>&mdash; Rohan Paul (@rohanpaul_ai) <a href=\"https:\/\/x.com\/rohanpaul_ai\/status\/2121370981403156931\" target=\"_blank\" rel=\"noreferrer noopener\">November 13, 2026 \u00b7 View on X<\/a><\/blockquote>\n\n\n\n<p class=\"wp-block-paragraph\">The launch post asked readers to comment &#8220;Reason&#8221; for early access and free credits, a format that inflates reply counts and says little about real-world use. i10x did not find a substantive Grok assessment of Reason on X; Grok&#8217;s replies only restated the launch claims.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">Critical analysis<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">Reason is tackling a real problem. Agent sessions are long, and providers discount cached tokens heavily, so a harness that keeps its prompt stable can save meaningful money. The design choices are sound and, unusually for a startup, documented with reproducible code.<\/p>\n\n\n\n<p class=\"wp-block-paragraph\">The marketing is another matter. &#8220;Up to 20x&#8221; and &#8220;22% on the same model&#8221; are not the same claim, and only the second is backed by a published, same-model comparison. Readers should also note that the comparison is Reason&#8217;s own reproduction of a third-party benchmark, not an official Artificial Analysis result, and that the frontier labs keep tuning their own harnesses (OpenAI&#8217;s Codex, Anthropic&#8217;s Claude Code) for caching too. A 22% edge today can shrink with one vendor update.<\/p>\n\n\n\n<h2 class=\"wp-block-heading\">What&#8217;s still open<\/h2>\n\n\n\n<ul class=\"wp-block-list\"><li>Which comparison produces the &#8220;up to 20x&#8221; figure. Reason has not published it in the materials i10x reviewed.<\/li><li>Independent results on Artificial Analysis&#8217;s current Coding Agent Index v2.0.<\/li><li>How Reason secures customer code and connected subscription credentials across cloud machines.<\/li><li>Funding: the founders say they are raising a seed round; no amount or investors have been announced.<\/li><\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\u270d\ufe0f About the analysis<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">This independent i10x analysis is based on Reason Machines&#8217; launch post, its Reason 1.1 benchmark write-up, its benchmark repository and product comparison page, and public posts on X. i10x did not run Reason or reproduce the benchmarks.<\/p>\n\n\n\n<ul class=\"wp-block-list\"><li><a href=\"https:\/\/reasonmachines.com\/blog\/launch\" target=\"_blank\" rel=\"noreferrer noopener\">Reason Machines: Reason, the most efficient agent harness to maximize your AI usage<\/a><\/li><li><a href=\"https:\/\/reasonmachines.com\/blog\/introducing-reason-agent-1-1\" target=\"_blank\" rel=\"noreferrer noopener\">Reason Machines: Introducing Reason 1.1<\/a><\/li><li><a href=\"https:\/\/github.com\/reason-machines\/benchmark\" target=\"_blank\" rel=\"noreferrer noopener\">GitHub: reason-machines\/benchmark<\/a><\/li><li><a href=\"https:\/\/reasonmachines.com\/guides\/reason-vs-codex\" target=\"_blank\" rel=\"noreferrer noopener\">Reason Machines: Reason vs Codex (plans and benchmark caveats)<\/a><\/li><li><a href=\"https:\/\/x.com\/xadisingh\/status\/2121265353867284498\" target=\"_blank\" rel=\"noreferrer noopener\">Adi Singh on X, November 13, 2026<\/a><\/li><\/ul>\n\n\n\n<h2 class=\"wp-block-heading\">\ud83d\udd2d i10x Perspective<\/h2>\n\n\n\n<p class=\"wp-block-paragraph\">The most useful idea in this launch is not the 20x headline but the sentence &#8220;the harness sets the bill.&#8221; As models converge in quality, the money moves to the layer that decides how many tokens a task burns. That is good news for buyers, who now have a lever beyond switching models. It is also a warning: a harness that only wins on caching discounts lives at the mercy of provider pricing. Reason&#8217;s open benchmark code is the right move. Now it needs an independent number to match the marketing.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>YC startup Reason Machines went viral with a coding harness promising frontier results at up to 20x lower cost. Its published benchmark on GPT-6 Astra shows a smaller but real edge: 65.1 versus 62.5 for Codex at 22% lower cost per solved task, thanks to cache-friendly prompts and parallel tool calls.<\/p>\n","protected":false},"author":2,"featured_media":8928,"comment_status":"closed","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"paper_title":"","paper_url":"","pdf_url":"","doi":"","arxiv_id":"","authors":"","institutions":"","paper_date":"","venue":"","peer_reviewed":false,"code_url":"","data_url":"","funding_coi":"","figure_license":"","paper_license":"","replication":"","study_models":"","study_sample":"","footnotes":"","_yoast_wpseo_primary_category":"22","_nx_sec_severity":"","_nx_sec_affected":"","_nx_sec_status":"","_nx_sec_kicker":"","_nx_sec_type":"","_nx_sec_vendor":"","_nx_sec_fixed":"","_nx_sec_action":""},"categories":[14,5,22],"tags":[],"companies":[],"products":[],"class_list":["post-8929","post","type-post","status-publish","format-standard","has-post-thumbnail","hentry","category-agents","category-business","category-coding"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.7 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News<\/title>\n<meta name=\"description\" content=\"The Reason coding harness promises up to 20x lower agent costs. Its maker&#039;s own benchmark shows 22% lower cost per solved task than Codex.\" \/>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims\" \/>\n<meta property=\"og:locale\" content=\"en_US\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News\" \/>\n<meta property=\"og:description\" content=\"The Reason coding harness promises up to 20x lower agent costs. Its maker&#039;s own benchmark shows 22% lower cost per solved task than Codex.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims\" \/>\n<meta property=\"og:site_name\" content=\"i10X News\" \/>\n<meta property=\"article:published_time\" content=\"2026-11-14T07:28:50+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-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:image\" content=\"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/i10x-news-og-default-1200x630-1.png\" \/>\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=\"9 minutes\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"NewsArticle\",\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims\"},\"author\":{\"name\":\"Christopher Ort\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#\\\/schema\\\/person\\\/d410bb7ee72febefca2c9c369939219a\"},\"headline\":\"Reason Coding Harness Promises 20x Lower Costs. Its Own Benchmarks Show 22%\",\"datePublished\":\"2026-11-14T07:28:50+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims\"},\"wordCount\":1759,\"publisher\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#organization\"},\"image\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/reason-machines-coding-harness-cost-claims-title.webp\",\"articleSection\":[\"Agents\",\"Business\",\"Coding\"],\"inLanguage\":\"en-US\",\"dateModified\":\"2026-11-14T07:28:50+00:00\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims\",\"url\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims\",\"name\":\"Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#website\"},\"primaryImageOfPage\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#primaryimage\"},\"image\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#primaryimage\"},\"thumbnailUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/reason-machines-coding-harness-cost-claims-title.webp\",\"datePublished\":\"2026-11-14T07:28:50+00:00\",\"description\":\"The Reason coding harness promises up to 20x lower agent costs. Its maker's own benchmark shows 22% lower cost per solved task than Codex.\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#breadcrumb\"},\"inLanguage\":\"en-US\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims\"]}]},{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#primaryimage\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/reason-machines-coding-harness-cost-claims-title.webp\",\"contentUrl\":\"https:\\\/\\\/i10x.ai\\\/news\\\/wp-content\\\/uploads\\\/2026\\\/11\\\/reason-machines-coding-harness-cost-claims-title.webp\",\"width\":1344,\"height\":768,\"caption\":\"Image: AI-generated, i10X\"},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/i10xnews.kinsta.cloud\\\/reason-machines-coding-harness-cost-claims#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/i10x.ai\\\/news\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"Reason Coding Harness Promises 20x Lower Costs. Its Own Benchmarks Show 22%\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#website\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\",\"name\":\"i10X News\",\"description\":\"Stay updated with the latest news and announcements from i10X.\",\"publisher\":{\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#organization\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/i10x.ai\\\/news?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-US\"},{\"@type\":\"Organization\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#organization\",\"name\":\"i10X\",\"url\":\"https:\\\/\\\/i10x.ai\\\/news\\\/\",\"logo\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-US\",\"@id\":\"https:\\\/\\\/i10x.ai\\\/news\\\/#\\\/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:\\\/\\\/i10x.ai\\\/news\\\/#\\\/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:\\\/\\\/i10x.ai\\\/news\\\/#\\\/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\"},\"subjectOf\":[{\"@type\":\"CreativeWork\",\"url\":\"https:\\\/\\\/papers.ssrn.com\\\/abstract=7426699\",\"name\":\"The Style Penalty: How AI Resume-Writing Tools Shape Outcomes in Automated Hiring Screens\"},{\"@type\":\"CreativeWork\",\"url\":\"https:\\\/\\\/www.heise.de\\\/news\\\/KI-Chatbots-bewerten-KI-Lebenslaeufe-Claude-findet-sich-besser-als-GPT-11317531.html\",\"name\":\"KI-Chatbots bewerten KI-Lebensl\u00e4ufe: Claude findet sich besser als GPT\"},{\"@type\":\"CreativeWork\",\"url\":\"https:\\\/\\\/t3n.de\\\/news\\\/bewerbung-schreiben-ki-5-fehler-1747190\\\/\",\"name\":\"Bewerbung mit KI schreiben: Diese 5 Fehler solltet ihr laut Studien vermeiden\"}]}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News","description":"The Reason coding harness promises up to 20x lower agent costs. Its maker's own benchmark shows 22% lower cost per solved task than Codex.","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:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims","og_locale":"en_US","og_type":"article","og_title":"Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News","og_description":"The Reason coding harness promises up to 20x lower agent costs. Its maker's own benchmark shows 22% lower cost per solved task than Codex.","og_url":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims","og_site_name":"i10X News","article_published_time":"2026-11-14T07:28:50+00:00","og_image":[{"width":1344,"height":768,"url":"https:\/\/i10xnews.kinsta.cloud\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-title.webp","type":"image\/webp"}],"author":"Christopher Ort","twitter_card":"summary_large_image","twitter_image":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/10\/i10x-news-og-default-1200x630-1.png","twitter_creator":"@i10x_ai","twitter_site":"@i10x_ai","twitter_misc":{"Written by":"Christopher Ort","Est. reading time":"9 minutes"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"NewsArticle","@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#article","isPartOf":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims"},"author":{"name":"Christopher Ort","@id":"https:\/\/i10x.ai\/news\/#\/schema\/person\/d410bb7ee72febefca2c9c369939219a"},"headline":"Reason Coding Harness Promises 20x Lower Costs. Its Own Benchmarks Show 22%","datePublished":"2026-11-14T07:28:50+00:00","mainEntityOfPage":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims"},"wordCount":1759,"publisher":{"@id":"https:\/\/i10x.ai\/news\/#organization"},"image":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-title.webp","articleSection":["Agents","Business","Coding"],"inLanguage":"en-US","dateModified":"2026-11-14T07:28:50+00:00"},{"@type":"WebPage","@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims","url":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims","name":"Reason Coding Harness: 20x Cheaper Claim vs. 22% Data - i10X News","isPartOf":{"@id":"https:\/\/i10x.ai\/news\/#website"},"primaryImageOfPage":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#primaryimage"},"image":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#primaryimage"},"thumbnailUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-title.webp","datePublished":"2026-11-14T07:28:50+00:00","description":"The Reason coding harness promises up to 20x lower agent costs. Its maker's own benchmark shows 22% lower cost per solved task than Codex.","breadcrumb":{"@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#breadcrumb"},"inLanguage":"en-US","potentialAction":[{"@type":"ReadAction","target":["https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims"]}]},{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#primaryimage","url":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-title.webp","contentUrl":"https:\/\/i10x.ai\/news\/wp-content\/uploads\/2026\/11\/reason-machines-coding-harness-cost-claims-title.webp","width":1344,"height":768,"caption":"Image: AI-generated, i10X"},{"@type":"BreadcrumbList","@id":"https:\/\/i10xnews.kinsta.cloud\/reason-machines-coding-harness-cost-claims#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/i10x.ai\/news"},{"@type":"ListItem","position":2,"name":"Reason Coding Harness Promises 20x Lower Costs. Its Own Benchmarks Show 22%"}]},{"@type":"WebSite","@id":"https:\/\/i10x.ai\/news\/#website","url":"https:\/\/i10x.ai\/news","name":"i10X News","description":"Stay updated with the latest news and announcements from i10X.","publisher":{"@id":"https:\/\/i10x.ai\/news\/#organization"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/i10x.ai\/news?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-US"},{"@type":"Organization","@id":"https:\/\/i10x.ai\/news\/#organization","name":"i10X","url":"https:\/\/i10x.ai\/news\/","logo":{"@type":"ImageObject","inLanguage":"en-US","@id":"https:\/\/i10x.ai\/news\/#\/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:\/\/i10x.ai\/news\/#\/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:\/\/i10x.ai\/news\/#\/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"},"subjectOf":[{"@type":"CreativeWork","url":"https:\/\/papers.ssrn.com\/abstract=7426699","name":"The Style Penalty: How AI Resume-Writing Tools Shape Outcomes in Automated Hiring Screens"},{"@type":"CreativeWork","url":"https:\/\/www.heise.de\/news\/KI-Chatbots-bewerten-KI-Lebenslaeufe-Claude-findet-sich-besser-als-GPT-11317531.html","name":"KI-Chatbots bewerten KI-Lebensl\u00e4ufe: Claude findet sich besser als GPT"},{"@type":"CreativeWork","url":"https:\/\/t3n.de\/news\/bewerbung-schreiben-ki-5-fehler-1747190\/","name":"Bewerbung mit KI schreiben: Diese 5 Fehler solltet ihr laut Studien vermeiden"}]}]}},"_links":{"self":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/8929","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=8929"}],"version-history":[{"count":0,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/posts\/8929\/revisions"}],"wp:featuredmedia":[{"embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media\/8928"}],"wp:attachment":[{"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/media?parent=8929"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/categories?post=8929"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/tags?post=8929"},{"taxonomy":"company","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/companies?post=8929"},{"taxonomy":"product","embeddable":true,"href":"https:\/\/i10x.ai\/news\/wp-json\/wp\/v2\/products?post=8929"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}