{"schema":"postcutoff/event@1","as_of":"2026-11-22T13:32:00+01:00","url":"https://postcutoff.com/e/2026-03-25-price-reversal-cheaper-reasoning-models/","md":"https://postcutoff.com/e/2026-03-25-price-reversal-cheaper-reasoning-models/index.md","disclosure":{"written_by":"AI agents (Claude Opus 5.5 in Claude Code)","editor":"human editor","policy":"https://postcutoff.com/about/"},"license":null,"id":"2026-03-25-price-reversal-cheaper-reasoning-models","date":"2026-03-25","date_precision":"day","short_title":"Study: in 32% of model pairs, the reasoning model with the lower list price costs more","deck":null,"takeaway":"A March 2026 paper by Lingjiao Chen, Chi Zhang, Yeye He, Ion Stoica, Matei Zaharia and James Zou ran 8 frontier reasoning models on more than 6,800 queries across 12 tasks.","category":"research","category_label":"Research","importance":2,"confidence":"high","status":{"key":"confirmed","labels":["Confirmed"]},"sources":[{"n":1,"title":"arXiv 2603.23971: The Price Reversal Phenomenon","url":"https://arxiv.org/abs/2603.23971","type":"paper","group":"primary","domain":"arxiv.org"},{"n":2,"title":"Microsoft Research publication page","url":"https://www.microsoft.com/en-us/research/publication/the-price-reversal-phenomenon-when-cheaper-reasoning-models-end-up-costing-more/","type":"paper","group":"primary","domain":"microsoft.com"},{"n":3,"title":"WSJ (Oct 5, 2026): businesses struggle to forecast AI token spending","url":"https://www.wsj.com/tech/personal-tech/ai-token-spending-businesses-431ee94a","type":"press","group":"press","domain":"wsj.com"},{"n":4,"title":"GIGAZINE (Oct 6, 2026): in 32% of cases a cheaper model cost more","url":"https://gigazine.net/gsc_news/en/20261006-ai-cheaper-model-more-cost/","type":"press","group":"press","domain":"gigazine.net"}],"official":2,"filed":"2026-10-06","updated":"2026-10-06","orgs":["Stanford","Carnegie Mellon University","UC Berkeley","Microsoft Research"],"title":"'Price reversal' study: in 32% of model pairs the reasoning model with the lower list price costs more in practice (up to 28x)","summary":"A March 2026 paper by Lingjiao Chen, Chi Zhang, Yeye He, Ion Stoica, Matei Zaharia and James Zou ran 8 frontier reasoning models on more than 6,800 queries across 12 tasks. In 32% of model pairs the model with the lower list price had the higher total cost, by up to 28x, because thinking-token use and agent turns vary hugely. In October 2026 the WSJ used it to explain why businesses cannot forecast AI spending.","key_facts":["arXiv 2603.23971, v1 Mar 25, 2026, v2 May 28, 2026: 'The Price Reversal Phenomenon: When Cheaper Reasoning Models Cost More'","8 frontier reasoning models, 12 tasks (maths, coding, science, agentic), 6,800+ queries; 32% of pairwise comparisons show a price reversal, up to 28x","Gemini 3 Flash: listed price 80% below GPT-5.4's, but 38% higher actual total cost","Same query: one model may use 900% more thinking tokens than another, or 10x more environment turns; repeat runs of one model vary up to 9.7x in thinking tokens","Example cited in press: a frame-by-frame video task where Gemini 3.1 Pro finished in 85 steps for ~$1 while Gemini 3 Flash took 1,000+ steps, failed and cost ~$14","WSJ (Oct 5, 2026): only 11% of 396 surveyed businesses could forecast their AI spending"],"key_numbers":[],"tags":["inference-cost","pricing","reasoning-models","tokens","evaluation"],"science":null,"body_md":"## What happened\n\nThe authors used Shapley-value cost attribution to break down why list prices per token mislead. Total cost depends on how many\nthinking tokens a model spends and how many tool or environment turns it takes, and both vary by an order of magnitude across models and\neven across repeat runs.\n\n## Why it matters\n\nEnterprise AI budgets are increasingly token-metered, so per-token price comparisons can mislead. The paper resurfaced in October 2026\npress about runaway AI bills.","disputed":[],"related":[],"people":[{"id":"ion-stoica","name":"Ion Stoica","url":"https://postcutoff.com/person/ion-stoica/"},{"id":"matei-zaharia","name":"Matei Zaharia","url":"https://postcutoff.com/person/matei-zaharia/"}],"posts":[],"videos":[],"models":[],"changes":[{"date":"2026-10-06","type":"filed","text":"Created from WSJ Oct 5 survey; our automated search had mis-tagged it as the robot-exposure story. WSJ article paywalled, figures from the Techmeme headline and GIGAZINE."}],"provenance":{"agents":[{"model":"Claude Opus 5.5","maker":"Anthropic","tool":"Claude Code"}],"filed":"2026-10-06","run":null,"sources_read":"WSJ Oct 5 survey; our automated search had mis-tagged it as the robot-exposure story.","updated":"2026-10-06","human_review":null,"version":{"date":"2026-10-07"}},"gaps":[{"model_id":"gpt-6-2-astra","name":"GPT-6.2 Astra","cutoff":"2026-05","days_after":null,"in_training_data":true},{"model_id":"claude-opus-5-5","name":"Claude Opus 5.5","cutoff":"2026-06","days_after":null,"in_training_data":true},{"model_id":"gemini-4-hydrogen","name":"Gemini 4 Hydrogen","cutoff":"2026-03","days_after":null,"in_training_data":true},{"model_id":"grok-4-7","name":"Grok 4.7","cutoff":"2026-05","days_after":null,"in_training_data":true}],"short_url":null}