Thinking Machines Lab signs a $65 million inference deal with Crusoe, a GPU cloud and energy-efficient compute provider.

Crusoe, a vertically integrated AI infrastructure provider, and Thinking Machines Lab have signed a $65 million annual agreement to power inference for the lab's open models on Crusoe Cloud. Through Crusoe Managed Inference, Thinking Machines Lab will operate production workloads—including its Inkling models, GLM 5.2 and 5.3, and fine-tuned variants—on a dedicated deployment of NVIDIA HGX B200 systems connected with NVIDIA Quantum-2 InfiniBand networking, designed for high-throughput, cost-efficient serving at scale.

"Crusoe Managed Inference took us from evaluation to production quickly and met the standards we hold our own systems to, giving us the scale and economics to reinvest in the research at the core of what we do," said Myle Ott, ML Infra Lead at Thinking Machines Lab.

Crusoe will run and support the cluster as a Tailored Deployment—a dedicated, benchmarked, SLA-backed endpoint—allowing Thinking Machines Lab to achieve the price-performance and headroom needed to scale without managing its own inference infrastructure. The companies are also exploring batch inference for large-scale synthetic data generation, positioning inference as a flywheel for research advancement.

"Thinking Machines Lab has a very sophisticated engineering team that expects partners to meet their high bar as they grow," said Erwan Menard, SVP of Product Management at Crusoe Cloud. "Our team of leading AI engineers built Crusoe Managed Inference to deliver cost-efficient, reliable infrastructure, inference, and model lifecycle tooling as-a-service, enabling innovative companies to run their favorite models in production at scale."

The partnership brings Crusoe Managed Inference's contracted annual recurring revenue to more than $100 million, less than a year after its launch.