[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"branding":3,"analytics":7,"article-auctions-reveal-how-overpriced-fixed-llm-tokens-really-are":10,"sections":35},{"siteName":4,"siteTagline":5,"publisherName":4,"contactEmail":6},"The Revision","Tech news, decoded.","editor@therevision.news",{"gaMeasurementId":8,"adsenseClientId":9},"G-ZW2MV82GYR","ca-pub-8533917693782264",{"article":11},{"id":12,"slug":13,"title":14,"dek":15,"body_md":16,"tags_json":17,"published_at":18,"created_at":19,"updated_at":20,"status":21,"review_note":22,"review_notes":23,"image_url":22,"persona_id":22,"persona_name":22,"section":24,"tags":25,"sources":30,"feedback":34,"feedback_at":22,"cost_usd":34,"total_tokens":34},7547,"auctions-reveal-how-overpriced-fixed-llm-tokens-really-are","Auctions Reveal How Overpriced Fixed LLM Tokens Really Are","A new procurement platform makes AI providers bid for queries and finds fixed per-token pricing can overcharge users by up to 71 percent.","Most AI API pricing is a fixed number per token, set by the provider and taken or left by the customer. New research suggests that number is often just wrong.\n\nResearchers built a procurement platform that routes each query through a reverse second-price auction, the same mechanism eBay-style sellers face when bidding to win a contract at the lowest price. Model providers bid what they estimate a query actually costs to serve, and the platform learns each provider's quality over time, funneling queries to whichever one is cheapest while still clearing a quality bar the user sets. The team tested it with Llama and Qwen family models on math-reasoning and question-answering benchmarks. Depending on the task and the quality threshold demanded, the cheapest qualifying provider undercut the going fixed rate by anywhere from 10 percent to 71 percent.\n\n\n\nThat spread is the real story. A 71 percent gap is not rounding error, it is evidence that flat per-token pricing bakes in a lot of margin that competition would otherwise strip out. It is the same dynamic that turned cloud compute from list-price contracts into spot markets and reserved-instance discounts. If LLM inference follows that path, the platforms sitting between providers and users, not the model labs themselves, stand to gain the most leverage.\n\nThis is a lab prototype, not a live marketplace, and it only proves the inefficiency exists under benchmark conditions. Whether providers would tolerate a bidding war on their own margins, rather than just on router traffic, is the open question nobody in this paper had to answer.","[\"llm pricing\",\"ai infrastructure\",\"model routing\",\"auctions\"]","2026-09-24T04:00:00.000Z","2026-09-24T05:41:19.326Z","2026-09-24T05:41:23.041Z","published",null,[],"ai",[26,27,28,29],"llm pricing","ai infrastructure","model routing","auctions",[31],{"name":32,"url":33},"arXiv cs.AI","https:\u002F\u002Farxiv.org\u002Fabs\u002F2609.28322",0,{"sections":36},[37,40,44,49,54,59,64,69,74,79,84,89,94,99],{"name":38,"slug":24,"count":39,"latest_published_at":18},"AI",4424,{"name":41,"slug":42,"count":43,"latest_published_at":18},"Security","security",724,{"name":45,"slug":46,"count":47,"latest_published_at":48},"Policy","policy",380,"2026-09-23T22:53:43.000Z",{"name":50,"slug":51,"count":52,"latest_published_at":53},"Deals","deals",221,"2026-09-24T08:00:00.000Z",{"name":55,"slug":56,"count":57,"latest_published_at":58},"Hardware","hardware",173,"2026-09-23T23:42:10.000Z",{"name":60,"slug":61,"count":62,"latest_published_at":63},"Science","science",136,"2026-09-24T09:00:00.000Z",{"name":65,"slug":66,"count":67,"latest_published_at":68},"Consumer Tech","consumer-tech",116,"2026-09-24T00:51:49.000Z",{"name":70,"slug":71,"count":72,"latest_published_at":73},"Software","software",85,"2026-09-23T20:00:00.000Z",{"name":75,"slug":76,"count":77,"latest_published_at":78},"Dev Tools","dev-tools",79,"2026-09-22T22:21:13.000Z",{"name":80,"slug":81,"count":82,"latest_published_at":83},"Startups","startups",66,"2026-09-23T17:28:38.000Z",{"name":85,"slug":86,"count":87,"latest_published_at":88},"Gaming","gaming",45,"2026-09-22T15:35:06.000Z",{"name":90,"slug":91,"count":92,"latest_published_at":93},"General","general",43,"2026-09-21T23:48:56.000Z",{"name":95,"slug":96,"count":97,"latest_published_at":98},"Reviews","reviews",27,"2026-09-22T13:00:00.000Z",{"name":100,"slug":101,"count":102,"latest_published_at":103},"How-To","how-to",6,"2026-06-16T09:00:00.000Z"]