SpaceX has disclosed in a regulatory filing a binding commitment for about $40 billion in financing with Apollo Global Management for AI chips and related data center equipment. Apollo will buy the hardware and lease it to SpaceX for about five to 5.25 years, with the deal expected to close by year-end and leases starting in the first half of 2027.
SpaceX has signed a roughly $40 billion GPU and equipment financing lease agreement with Apollo Global Management to procure Nvidia's next-generation Vera Rubin chips for its "superintelligent future" compute buildout. Equipment will be delivered in batches with lease terms of about five to 5.25 years per batch, starting in the first half of 2027.
At the GTC Berlin keynote on October 23, 2026, Nvidia CEO Jensen Huang demonstrated Microduck robots and the video went viral. The following day, Hugging Face's Clem Delangue said preorders surpassed 20,000, with deliveries expected before Christmas. Microduck is an open-source robot produced by Pollen Robotics, which Hugging Face acquired in 2025; Nvidia acquired Hugging Face in September.
Microsoft and Hugging Face released ThinkingBox-Training, which converts executable checks of database state into reinforcement learning rewards for training agents. On Qwen3.8-27B, average success rose from 51.70% to 60.78%, and the number of tasks completed correctly on all attempts rose from 38 to 107.
NVIDIA announced SimReady Central Library, a free collection of over 2,000 simulation-ready OpenUSD assets on Hugging Face for physical AI training and AI factory digital twins, covering robot arms, grippers, warehouse equipment, racks, and cooling infrastructure, with OEMs able to validate and distribute assets to developers.
SpaceX CFO Brett Johnson said at an earnings call that the company's superintelligent infrastructure segment has strong growth prospects, compute market demand is strong, payback for newly deployed compute capacity is under one year, and per-watt compute prices are slightly above the upper end of last quarter's $30-50 range. Musk added that superintelligent compute shortage will persist long-term, with payback potentially shortening to six months in 2027.
Musk said at a SpaceX conference call that Terafab, a chip venture jointly owned by SpaceX and Tesla, plans to reach volume production in Q3 2028, producing 1 terawatt of compute annually for AI and robotics. He also disclosed an internal target to produce a SRAM test chip at the Austin base around May or June next year.
At SpaceX's Q3 earnings call, Elon Musk confirmed that SpaceX and Tesla have formed special committees to discuss a potential merger. No agreement has been reached yet, and any deal would require shareholder approval. Musk said he personally thinks a merger would be good but recused himself as an interested party. Analysts estimate a probability above 80%.
SpaceX reported $14.6 billion in third-quarter revenue, with its AI business (renamed Super Intelligence) contributing $7.5 billion, including Cursor, Grok, and a 1.7 GW neocloud business. The AI business posted $431 million in profit. Capital expenditures were $19.6 billion, and the CFO expects $35-40 billion in new capital investments in the fourth quarter.
Hugging Face Chief Policy Officer Irene Solaiman said in an interview at TechCrunch Disrupt that AI model evaluation methods remain immature, citing issues with reproducibility, comparability across benchmarks, and standardization of the embedded evaluation ecosystem, while emphasizing the importance of open weights for reproducible evaluations.