TensorWave Raises $350 Million to Expand AMD-Powered AI Infrastructure
TensorWave is making a major move in the race to deliver high-performance AI compute. The company has announced $350 million in Series B funding aimed at accelerating the deployment of AMD-powered artificial intelligence infrastructure, with a strong focus on the AMD Instinct MI355X GPU accelerator.
The investment comes as demand for AI compute continues to surge across industries. Companies building large language models, generative AI tools, and high-throughput inference systems are facing growing pressure to secure reliable GPU capacity. TensorWave is positioning itself as an alternative for organizations that need scalable AI infrastructure without being tied to a single closed ecosystem.
The new funding will help TensorWave expand its fleet of AMD Instinct GPUs and support larger, more memory-intensive workloads. According to the company, its upcoming deployments are designed for demanding AI use cases such as LLM training, production inference, and advanced generative AI applications.
TensorWave’s focus on AMD Instinct MI355X GPUs signals growing industry interest in AMD’s AI accelerator roadmap. As AI models become larger and more complex, memory capacity and bandwidth are becoming just as important as raw compute power. The MI355X is expected to play a key role in workloads that require large-scale parallel processing and efficient handling of massive datasets.
TensorWave CEO and co-founder Darrick Horton said the next stage of artificial intelligence will depend on access to enough compute power to move projects from experimentation into real-world production. He noted that enterprises increasingly need infrastructure that offers memory capacity, performance, and flexibility while avoiding vendor lock-in.
The company is already gaining traction with AI-focused businesses. Fireworks AI and Luma AI are among the firms using TensorWave’s AMD-based infrastructure to power large-scale generative AI workloads and production inference platforms. This reflects a broader market shift as AI companies look for more available and cost-effective compute options.
TensorWave has been building its AMD-powered AI platform rapidly. Since its Series A funding round in May 2025, the company has grown into one of the largest operators of AMD-based AI training clusters in the United States. It currently has 8,192 AMD Instinct MI325X AI GPUs online, while new deployments of AMD Instinct MI355X GPUs are underway across multiple North American data centers.
The company has also secured more than 2 gigawatts of long-term AI data center capacity. That power footprint is intended to support rising demand from enterprise customers, research organizations, and AI-native companies that need scalable infrastructure for training and inference.
TensorWave’s strategy centers on offering an open, AMD-powered foundation for production AI. This approach is especially relevant as the global GPU market remains tight and many organizations continue to search for alternatives to capacity-constrained supply chains. By expanding access to AMD accelerators, TensorWave aims to give customers more choice in how they build and scale AI systems.
Investor confidence in the company appears to be tied to both the rising demand for AI compute and TensorWave’s ability to deploy infrastructure quickly. Ross Laser, co-founder and president of Magnetar, said the AI infrastructure race has created urgent demand for providers that can deliver capacity at speed while maintaining reliability. He highlighted TensorWave’s partnership with AMD and its execution as key reasons the company is well positioned in the market.
The timing of the investment is also important for AMD. The company is preparing to ship its next-generation Instinct MI455X GPU accelerators, which are expected to deliver higher compute performance, improved efficiency, and lower total cost of ownership for AI data centers. These upcoming chips are expected to compete with future NVIDIA data center platforms as cloud providers, AI labs, and enterprises continue expanding next-generation AI infrastructure.
For TensorWave, the $350 million Series B round represents more than just a funding milestone. It strengthens the company’s push to become a major provider of AMD-based AI compute at a time when demand for training and inference capacity is accelerating worldwide.
As artificial intelligence moves deeper into production environments, infrastructure providers that can offer scalable, flexible, and high-memory GPU clusters will become increasingly important. TensorWave is betting that AMD’s Instinct GPU lineup, combined with large-scale data center capacity, can give AI builders the performance and availability they need to compete in the next wave of AI development.
