{"id":158765,"date":"2023-02-22T20:26:40","date_gmt":"2023-02-23T02:26:40","guid":{"rendered":"https:\/\/lifeboat.com\/blog\/2023\/02\/openais-foundry-will-let-customers-buy-dedicated-compute-to-run-its-ai-models"},"modified":"2023-02-22T20:26:40","modified_gmt":"2023-02-23T02:26:40","slug":"openais-foundry-will-let-customers-buy-dedicated-compute-to-run-its-ai-models","status":"publish","type":"post","link":"https:\/\/lifeboat.com\/blog\/2023\/02\/openais-foundry-will-let-customers-buy-dedicated-compute-to-run-its-ai-models","title":{"rendered":"OpenAI\u2019s Foundry will let customers buy dedicated compute to run its AI models"},"content":{"rendered":"<p><a class=\"aligncenter blog-photo\" href=\"https:\/\/lifeboat.com\/blog.images\/openais-foundry-will-let-customers-buy-dedicated-compute-to-run-its-ai-models2.jpg\"><\/a><\/p>\n<p>OpenAI is quietly launching a new developer platform that lets customers run the company\u2019s newer machine learning models, like <a href=\"https:\/\/techcrunch.com\/2022\/12\/01\/while-anticipation-builds-for-gpt-4-openai-quietly-releases-gpt-3-5\/\">GPT-3.5<\/a>, on dedicated capacity. In screenshots of documentation published to Twitter by <a href=\"https:\/\/twitter.com\/transitive_bs\/status\/1628118176524533760\/photo\/1\" target=\"_blank\" rel=\"noopener\">users<\/a> with early access, OpenAI describes the forthcoming offering, called Foundry, as \u201cdesigned for cutting-edge customers running larger workloads.\u201d<\/p>\n<p>\u201c[Foundry allows] inference at scale with full control over the model configuration and performance profile,\u201d the documentation reads. We\u2019ve reached out to OpenAI to confirm the veracity.<\/p>\n<p>If the screenshots are to be believed, Foundry \u2014 whenever it launches \u2014 will deliver a \u201cstatic allocation\u201d of compute capacity (perhaps on Azure, OpenAI\u2019s preferred public cloud platform) dedicated to a single customer. Users will be able to monitor specific instances with the same tools and dashboards that OpenAI uses to build and optimize models. In addition, Foundry will provide some level of version control, letting customers decide whether or not to upgrade to newer model releases, as well as \u201cmore robust\u201d fine-tuning for OpenAI\u2019s latest models.<\/p>\n","protected":false},"excerpt":{"rendered":"<p>OpenAI is quietly launching a new developer platform that lets customers run the company\u2019s newer machine learning models, like GPT-3.5, on dedicated capacity. In screenshots of documentation published to Twitter by users with early access, OpenAI describes the forthcoming offering, called Foundry, as \u201cdesigned for cutting-edge customers running larger workloads.\u201d \u201c[Foundry allows] inference at scale [\u2026]<\/p>\n","protected":false},"author":609,"featured_media":0,"comment_status":"open","ping_status":"closed","sticky":false,"template":"","format":"standard","meta":{"footnotes":""},"categories":[6],"tags":[],"class_list":["post-158765","post","type-post","status-publish","format-standard","hentry","category-robotics-ai"],"_links":{"self":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/158765","targetHints":{"allow":["GET"]}}],"collection":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts"}],"about":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/types\/post"}],"author":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/users\/609"}],"replies":[{"embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/comments?post=158765"}],"version-history":[{"count":0,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/posts\/158765\/revisions"}],"wp:attachment":[{"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/media?parent=158765"}],"wp:term":[{"taxonomy":"category","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/categories?post=158765"},{"taxonomy":"post_tag","embeddable":true,"href":"https:\/\/lifeboat.com\/blog\/wp-json\/wp\/v2\/tags?post=158765"}],"curies":[{"name":"wp","href":"https:\/\/api.w.org\/{rel}","templated":true}]}}