TECH NEWS

SpaceX’s AI business is beating expectations

https://techcrunch.com/wp-content/uploads/2026/06/elon-nasdaq-spacex-ipo.jpg?resize=1200,801

1:40 PM PDT · October 20, 2026

SpaceX’s data centers and Grok agents have been more profitable than investors expected in second quarter of the year, as the company leaves its roots as a space company behind to focus on artificial intelligence, which now makes up 52% of its income.

Elon Musk’s deep conglomerate reported on Tuesday $14.6 billion in third-quarter revenue. That number was driven by growth in all segments, but especially it’s renamed “Super Intelligence” business line, which includes the recently-acquired Cursor coding tool, Grok bot, and most importantly, the company’s burgeoning neocloud business, which now boasts 1.7 GW of compute.

That number was also boosted by about $1 billion in income related to the recovery of bitcoin prices and an unspecified related-party income that might be attributed to a SpaceX investment in another company.

The growing revenue is balanced against total capital expenditures of $19.6 billion, as...

Copyright of this story solely belongs to techcrunch.com. To see the full text click HERE

Read more

https://d2908q01vomqb2.cloudfront.net/f1f836cb4ea6efb2a0b1b99f41ad8b103eff4b59/2026/10/14/ML-21817-featured-image.png

Beyond the KV cache: Training and serving i14’s Laplace 3B linear attention model on AWS Trainium2 with native PyTorch | Amazon Web Services

Most large language models (LLMs) use transformer architectures with softmax attention. During inference, their key-value (KV) cache grows with the conversation or document context. This can reduce concurrency, increase serving cost, and force agentic applications to discard useful history. i14 designed Laplace 3B as a linear attention alternative. Its recurrent