{"id":1373,"date":"2026-07-12T18:00:00","date_gmt":"2026-07-12T17:00:00","guid":{"rendered":"https:\/\/wade.one\/blog\/?p=1373"},"modified":"2026-07-12T18:00:00","modified_gmt":"2026-07-12T17:00:00","slug":"gpt-5-6-makes-model-choice-about-cost-again","status":"publish","type":"post","link":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/","title":{"rendered":"GPT-5.6 Makes Model Choice About Cost Again"},"content":{"rendered":"<p>GPT-5.6 is interesting, but the part I care about is not another benchmark saying the newest model is the cleverest. It is that OpenAI has made the tradeoff explicit. Sol, Terra, and Luna are three different points on the capability and cost curve, and that is much closer to how real workloads need to be treated.<\/p>\n<p>In the <a href=\"https:\/\/openai.com\/index\/gpt-5-6\/\">GPT-5.6 announcement<\/a>, OpenAI positions Sol as the flagship, Terra as the balanced option, and Luna as the fastest and cheapest. API pricing follows that split: Sol is $5 per million input tokens and $30 per million output tokens, Terra is $2.50 and $15, while Luna is $1 and $6. OpenAI also says Terra is competitive with GPT-5.5 and Luna can get close to it for substantially less money.<\/p>\n<p>That should change the default question from &#8220;which model is best?&#8221; to &#8220;what does this task actually need?&#8221; A difficult migration, security investigation, or long-running Codex job may justify Sol. Routine classification, extraction, summaries, and predictable tool calls probably do not. The sensible setup is to test all three on your own small evaluation set and route work by evidence rather than by model prestige.<\/p>\n<p>The annoying part is that this creates another decision teams have to own. The useful part is that the decision can now save real money. I would rather have clear tiers and measure them than pay flagship prices for every background job because nobody wanted to choose.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>GPT-5.6&#8217;s Sol, Terra, and Luna tiers make model choice a practical cost decision instead of a hunt for one universal default.<\/p>\n","protected":false},"author":0,"featured_media":0,"comment_status":"closed","ping_status":"open","sticky":false,"template":"","format":"standard","meta":{"_jetpack_newsletter_access":"","_jetpack_dont_email_post_to_subs":false,"_jetpack_newsletter_tier_id":0,"_jetpack_memberships_contains_paywalled_content":false,"_jetpack_feature_clip_id":0,"_jetpack_memberships_contains_paid_content":false,"footnotes":"","jetpack_publicize_message":"","jetpack_publicize_feature_enabled":true,"jetpack_social_post_already_shared":true,"jetpack_social_options":{"image_generator_settings":{"template":"highway","default_image_id":0,"font":"","enabled":false},"version":2},"jetpack_post_was_ever_published":false},"categories":[26],"tags":[116,194,196,141,195],"class_list":["post-1373","post","type-post","status-publish","format-standard","hentry","category-ai","tag-codex","tag-gpt-5-6","tag-luna","tag-openai","tag-terra"],"yoast_head":"<!-- This site is optimized with the Yoast SEO plugin v28.0 - https:\/\/yoast.com\/product\/yoast-seo-wordpress\/ -->\n<title>GPT-5.6 Makes Model Choice About Cost Again - wade.one<\/title>\n<meta name=\"robots\" content=\"index, follow, max-snippet:-1, max-image-preview:large, max-video-preview:-1\" \/>\n<link rel=\"canonical\" href=\"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/\" \/>\n<meta property=\"og:locale\" content=\"en_GB\" \/>\n<meta property=\"og:type\" content=\"article\" \/>\n<meta property=\"og:title\" content=\"GPT-5.6 Makes Model Choice About Cost Again - wade.one\" \/>\n<meta property=\"og:description\" content=\"GPT-5.6&#039;s Sol, Terra, and Luna tiers make model choice a practical cost decision instead of a hunt for one universal default.\" \/>\n<meta property=\"og:url\" content=\"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/\" \/>\n<meta property=\"og:site_name\" content=\"wade.one\" \/>\n<meta property=\"article:published_time\" content=\"2026-07-12T17:00:00+00:00\" \/>\n<meta property=\"og:image\" content=\"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/Wade-Logo-cropped.png\" \/>\n\t<meta property=\"og:image:width\" content=\"1518\" \/>\n\t<meta property=\"og:image:height\" content=\"1506\" \/>\n\t<meta property=\"og:image:type\" content=\"image\/png\" \/>\n<meta name=\"twitter:card\" content=\"summary_large_image\" \/>\n<meta name=\"twitter:creator\" content=\"@wadewomersley\" \/>\n<meta name=\"twitter:site\" content=\"@wadewomersley\" \/>\n<meta name=\"twitter:label1\" content=\"Estimated reading time\" \/>\n\t<meta name=\"twitter:data1\" content=\"1 minute\" \/>\n<script type=\"application\/ld+json\" class=\"yoast-schema-graph\">{\"@context\":\"https:\\\/\\\/schema.org\",\"@graph\":[{\"@type\":\"Article\",\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/#article\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/\"},\"author\":{\"name\":\"\",\"@id\":\"\"},\"headline\":\"GPT-5.6 Makes Model Choice About Cost Again\",\"datePublished\":\"2026-07-12T17:00:00+00:00\",\"mainEntityOfPage\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/\"},\"wordCount\":239,\"publisher\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/#\\\/schema\\\/person\\\/8b4739f8f8bb2cff5d792d4b8779fcc3\"},\"keywords\":[\"codex\",\"gpt-5-6\",\"luna\",\"openai\",\"terra\"],\"articleSection\":[\"AI\"],\"inLanguage\":\"en-GB\"},{\"@type\":\"WebPage\",\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/\",\"url\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/\",\"name\":\"GPT-5.6 Makes Model Choice About Cost Again - wade.one\",\"isPartOf\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/#website\"},\"datePublished\":\"2026-07-12T17:00:00+00:00\",\"breadcrumb\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/#breadcrumb\"},\"inLanguage\":\"en-GB\",\"potentialAction\":[{\"@type\":\"ReadAction\",\"target\":[\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/\"]}]},{\"@type\":\"BreadcrumbList\",\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/2026\\\/07\\\/12\\\/gpt-5-6-makes-model-choice-about-cost-again\\\/#breadcrumb\",\"itemListElement\":[{\"@type\":\"ListItem\",\"position\":1,\"name\":\"Home\",\"item\":\"https:\\\/\\\/wade.one\\\/blog\\\/\"},{\"@type\":\"ListItem\",\"position\":2,\"name\":\"GPT-5.6 Makes Model Choice About Cost Again\"}]},{\"@type\":\"WebSite\",\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/#website\",\"url\":\"https:\\\/\\\/wade.one\\\/blog\\\/\",\"name\":\"wade.one\",\"description\":\"wade womersley - york based software engineer\",\"publisher\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/#\\\/schema\\\/person\\\/8b4739f8f8bb2cff5d792d4b8779fcc3\"},\"potentialAction\":[{\"@type\":\"SearchAction\",\"target\":{\"@type\":\"EntryPoint\",\"urlTemplate\":\"https:\\\/\\\/wade.one\\\/blog\\\/?s={search_term_string}\"},\"query-input\":{\"@type\":\"PropertyValueSpecification\",\"valueRequired\":true,\"valueName\":\"search_term_string\"}}],\"inLanguage\":\"en-GB\"},{\"@type\":[\"Person\",\"Organization\"],\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/#\\\/schema\\\/person\\\/8b4739f8f8bb2cff5d792d4b8779fcc3\",\"name\":\"Wade Womersley\",\"image\":{\"@type\":\"ImageObject\",\"inLanguage\":\"en-GB\",\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/wp-content\\\/uploads\\\/2015\\\/02\\\/200px.png\",\"url\":\"https:\\\/\\\/wade.one\\\/blog\\\/wp-content\\\/uploads\\\/2015\\\/02\\\/200px.png\",\"contentUrl\":\"https:\\\/\\\/wade.one\\\/blog\\\/wp-content\\\/uploads\\\/2015\\\/02\\\/200px.png\",\"width\":202,\"height\":200,\"caption\":\"Wade Womersley\"},\"logo\":{\"@id\":\"https:\\\/\\\/wade.one\\\/blog\\\/wp-content\\\/uploads\\\/2015\\\/02\\\/200px.png\"}}]}<\/script>\n<!-- \/ Yoast SEO plugin. -->","yoast_head_json":{"title":"GPT-5.6 Makes Model Choice About Cost Again - wade.one","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:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/","og_locale":"en_GB","og_type":"article","og_title":"GPT-5.6 Makes Model Choice About Cost Again - wade.one","og_description":"GPT-5.6's Sol, Terra, and Luna tiers make model choice a practical cost decision instead of a hunt for one universal default.","og_url":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/","og_site_name":"wade.one","article_published_time":"2026-07-12T17:00:00+00:00","og_image":[{"width":1518,"height":1506,"url":"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/Wade-Logo-cropped.png","type":"image\/png"}],"twitter_card":"summary_large_image","twitter_creator":"@wadewomersley","twitter_site":"@wadewomersley","twitter_misc":{"Estimated reading time":"1 minute"},"schema":{"@context":"https:\/\/schema.org","@graph":[{"@type":"Article","@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/#article","isPartOf":{"@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/"},"author":{"name":"","@id":""},"headline":"GPT-5.6 Makes Model Choice About Cost Again","datePublished":"2026-07-12T17:00:00+00:00","mainEntityOfPage":{"@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/"},"wordCount":239,"publisher":{"@id":"https:\/\/wade.one\/blog\/#\/schema\/person\/8b4739f8f8bb2cff5d792d4b8779fcc3"},"keywords":["codex","gpt-5-6","luna","openai","terra"],"articleSection":["AI"],"inLanguage":"en-GB"},{"@type":"WebPage","@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/","url":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/","name":"GPT-5.6 Makes Model Choice About Cost Again - wade.one","isPartOf":{"@id":"https:\/\/wade.one\/blog\/#website"},"datePublished":"2026-07-12T17:00:00+00:00","breadcrumb":{"@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/#breadcrumb"},"inLanguage":"en-GB","potentialAction":[{"@type":"ReadAction","target":["https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/"]}]},{"@type":"BreadcrumbList","@id":"https:\/\/wade.one\/blog\/2026\/07\/12\/gpt-5-6-makes-model-choice-about-cost-again\/#breadcrumb","itemListElement":[{"@type":"ListItem","position":1,"name":"Home","item":"https:\/\/wade.one\/blog\/"},{"@type":"ListItem","position":2,"name":"GPT-5.6 Makes Model Choice About Cost Again"}]},{"@type":"WebSite","@id":"https:\/\/wade.one\/blog\/#website","url":"https:\/\/wade.one\/blog\/","name":"wade.one","description":"wade womersley - york based software engineer","publisher":{"@id":"https:\/\/wade.one\/blog\/#\/schema\/person\/8b4739f8f8bb2cff5d792d4b8779fcc3"},"potentialAction":[{"@type":"SearchAction","target":{"@type":"EntryPoint","urlTemplate":"https:\/\/wade.one\/blog\/?s={search_term_string}"},"query-input":{"@type":"PropertyValueSpecification","valueRequired":true,"valueName":"search_term_string"}}],"inLanguage":"en-GB"},{"@type":["Person","Organization"],"@id":"https:\/\/wade.one\/blog\/#\/schema\/person\/8b4739f8f8bb2cff5d792d4b8779fcc3","name":"Wade Womersley","image":{"@type":"ImageObject","inLanguage":"en-GB","@id":"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/200px.png","url":"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/200px.png","contentUrl":"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/200px.png","width":202,"height":200,"caption":"Wade Womersley"},"logo":{"@id":"https:\/\/wade.one\/blog\/wp-content\/uploads\/2015\/02\/200px.png"}}]}},"jetpack_publicize_connections":[],"jetpack_featured_media_url":"","jetpack_sharing_enabled":true,"jetpack-related-posts":[{"id":1294,"url":"https:\/\/wade.one\/blog\/2026\/04\/24\/chatgpt-5-5-and-codex-feel-like-a-real-step-up\/","url_meta":{"origin":1373,"position":0},"title":"ChatGPT 5.5 and Codex Feel Like a Real Step Up","author":"","date":"April 24, 2026","format":false,"excerpt":"After using GPT-5.5 in Codex, the improvement over GPT-5.4 feels less like a benchmark bump and more like better follow-through on real coding work.","rel":"","context":"In &quot;AI&quot;","block_context":{"text":"AI","link":"https:\/\/wade.one\/blog\/category\/ai\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":1103,"url":"https:\/\/wade.one\/blog\/2023\/03\/26\/google-bard-vs-chatgpt-a-quick-review\/","url_meta":{"origin":1373,"position":1},"title":"Google Bard vs ChatGPT &#8211; a quick review","author":"Wade","date":"March 26, 2023","format":false,"excerpt":"I am loving the evolution of AI in the last year: it's really developing rapidly and, if you still know how to rephrase your requests to work around limitations of the current implementations you can get a lot out of it. Google Bard and ChatGPT are two of the most\u2026","rel":"","context":"In &quot;AI&quot;","block_context":{"text":"AI","link":"https:\/\/wade.one\/blog\/category\/ai\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":1302,"url":"https:\/\/wade.one\/blog\/2026\/05\/14\/codex-is-moving-into-the-whole-development-workflow\/","url_meta":{"origin":1373,"position":2},"title":"Codex Is Moving Into the Whole Development Workflow","author":"Wade","date":"May 14, 2026","format":false,"excerpt":"OpenAI's latest Codex updates show the category moving beyond chat and autocomplete into review, terminals, browsers, and workflow context.","rel":"","context":"In &quot;AI&quot;","block_context":{"text":"AI","link":"https:\/\/wade.one\/blog\/category\/ai\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":410,"url":"https:\/\/wade.one\/blog\/2010\/01\/07\/snowy-britain-nasa-terra-image\/","url_meta":{"origin":1373,"position":3},"title":"Snowy Britain &#8211; NASA Terra image","author":"Wade","date":"January 7, 2010","format":false,"excerpt":"NASA uploaded an image today of Great Britain from space as of 11:50 UTC today, it's a truly remarkable picture showing just how much it has snowed here. For anyone not from the UK, snow like this has not hit Britain for over 30 years, hence all the hoo-haa about.\u2026","rel":"","context":"In &quot;News&quot;","block_context":{"text":"News","link":"https:\/\/wade.one\/blog\/category\/news\/"},"img":{"alt_text":"","src":"https:\/\/i0.wp.com\/wade.one\/blog\/wp-content\/uploads\/2016\/04\/GreatBritain.A2010007.1150.1km.jpg?resize=350%2C200&ssl=1","width":350,"height":200,"srcset":"https:\/\/i0.wp.com\/wade.one\/blog\/wp-content\/uploads\/2016\/04\/GreatBritain.A2010007.1150.1km.jpg?resize=350%2C200&ssl=1 1x, https:\/\/i0.wp.com\/wade.one\/blog\/wp-content\/uploads\/2016\/04\/GreatBritain.A2010007.1150.1km.jpg?resize=525%2C300&ssl=1 1.5x, https:\/\/i0.wp.com\/wade.one\/blog\/wp-content\/uploads\/2016\/04\/GreatBritain.A2010007.1150.1km.jpg?resize=700%2C400&ssl=1 2x"},"classes":[]},{"id":1333,"url":"https:\/\/wade.one\/blog\/2026\/06\/01\/codex-on-mobile-changes-the-shape-of-background-coding\/","url_meta":{"origin":1373,"position":4},"title":"Codex on Mobile Changes the Shape of Background Coding","author":"Wade","date":"June 1, 2026","format":false,"excerpt":"Codex on mobile is less about writing code on a phone and more about supervising background engineering work away from the desk.","rel":"","context":"In &quot;AI&quot;","block_context":{"text":"AI","link":"https:\/\/wade.one\/blog\/category\/ai\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]},{"id":1233,"url":"https:\/\/wade.one\/blog\/2026\/04\/24\/ai-coding-agents-are-becoming-normal-developer-infrastructure\/","url_meta":{"origin":1373,"position":5},"title":"AI Coding Agents Are Becoming Normal Developer Infrastructure","author":"Wade","date":"April 24, 2026","format":false,"excerpt":"AI coding agents are moving from novelty demos into normal developer infrastructure. The useful question now is how teams manage them properly.","rel":"","context":"In &quot;AI&quot;","block_context":{"text":"AI","link":"https:\/\/wade.one\/blog\/category\/ai\/"},"img":{"alt_text":"","src":"","width":0,"height":0},"classes":[]}],"_links":{"self":[{"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/posts\/1373","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/types\/post"}],"replies":[{"embeddable":true,"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/comments?post=1373"}],"version-history":[{"count":1,"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/posts\/1373\/revisions"}],"predecessor-version":[{"id":1382,"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/posts\/1373\/revisions\/1382"}],"wp:attachment":[{"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/media?parent=1373"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/categories?post=1373"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/wade.one\/blog\/wp-json\/wp\/v2\/tags?post=1373"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}