Tag
Briefings and ideas about Inference costs from AI podcast discussions and the research they cite, with links to the original sources.
AI should be sold by results delivered, not tokens, Moonshots panelists argued, citing Salesforce and OpenAI moves as they described them.
Updated Sep 2, 2026 IdeaPublishers could be paid each source's estimated Shapley value for AI agent use: roughly the cost of matching answer quality without it.
Updated Aug 25, 2026Modal CEO Erik Bernhardsson says GPUs are mostly sold as fixed multiyear blocks, while real demand rises and falls. After a $750 million raise, he argues that pooling thousands of customers lets Modal resell that capacity in flexible slices.
Tenex estimates its AI token bill at one to two times its employee costs. Alex Lieberman says falling prices make runaway spending less of a worry than six to eight months ago, and make it harder for enterprises to justify fine-tuning.
SF Compute CEO Evan Conrad says only about 5% of his company's capacity is on its order book, but customers recoup $25 per $100 on average by reselling. In his account, interruption terms, lead time and expected demand shape GPU prices.
A preprint reports GPT-6 Astra got 49 of 50 unpublished physics calculations right with web access, for about $2 each. On Moonshots, XPRIZE CEO Anousheh Ansari asked who will find new problems if AI solves the defined ones.
Peter Diamandis cited a fall from $2.36 to $0.38 per robot subtask between July and September and projected near-zero cost within a year. RoboCurve's published test instead compares two models at $1.14 and $0.90 per attempt.
Meta released Muse Wonder, a model built to remember users over time, on October 12. On Moonshots, Alex Wissner-Gross said it trails the frontier once cost counts; Salim Ismail said knowing users best may decide the next big AI contest.
Periodic Labs' Liam Fedus says the startup puts AI on lab instruments to capture richer data and automates bottlenecks one at a time. He calls full autonomy a non-goal and says humanoid robots would be a slower path to its aims.
Thomas Sohmers, co-founder of AI chip company Positron, says GPT-6 Astra and Opus 5.5 agents now run its chip tests on emulators in a closed loop. He says the company's token spend briefly exceeded its human salaries and peaked above $100,000 a day, and the agents have not yet had a novel design idea.
Positron CTO Thomas Sohmers said AI agents are multiplying demand for memory. He claimed Positron's first-generation product sustained 93% of theoretical memory bandwidth, against an average of 30–40% for NVIDIA GPUs when decoding with transformer models. Its next chip, Asimov, is planned for 2027.
Reflection AI's open-weight Beam is built for efficiency, by the company's own estimate. On Moonshots, Alexander Wissner-Gross said US open labs still lack a business model that would pay for frontier capability.
OutSystems CEO Woodson Martin says the company's AI token spending peaked in June and July, then fell below forecast after it built its own harness and model router. He argues most enterprise work doesn't need frontier models.
Yash Patil, whose company Applied Compute sells post-training, says teams should first improve prompts, context and tools, then retrain weights for their biggest workloads, tasks that change over time and judgment-heavy tasks.