TensorWave

Open-source is here to stay!

From CUDA portability to real-world performance on AMD GPUs, Michael Søndergaard, CEO of Spectral Compute shares why open, portable AI infrastructure matters more than ever. Watch the short interview from Beyond Summit 2026 on our YouTube channel: https://lnkd.in/gq46YUye

New blog: why multi-year GPU agreements have become a core part of production AI infrastructure strategy and what that says about the state of AI demand today. Read more: https://lnkd.in/d6F3SMyC

Today, we welcomed the next wave of AI builders with the Summer 2026 internship class here at TensorWave. From cloud infrastructure and AI systems to events, operations, and go-to-market initiatives, this cohort will get hands-on experience working inside one of the fastest-moving spaces in technology. Glad to have you here.

Spectral Compute used TensorWave’s AMD-native infrastructure to benchmark CUDA portability and performance on AMD Instinct™ MI355X GPUs. See how they did it - https://lnkd.in/gqTNQ6zu

So exciting to see TensorWave recognized by The Compute 100 as one of the companies helping shape the future of compute and making an impact on the industry! From day one, we’ve bet on an open, AMD-native approach to AI cloud infrastructure, and it’s exciting to see that vision gaining momentum across the industry. We’re just getting started. Check out our listing here: https://lnkd.in/gs8hqdst

You have to read this one. We just published a recap into how Wafer pushed AMD inference performance to a level that’s getting the entire ecosystem’s attention and the results are kind of wild. What makes this story interesting isn’t just the performance itself. It’s how they achieved it: systems-level optimization, smart inference tuning, and a belief that AMD can compete at the very highest tier. Proud this work was powered on TensorWave’s AMD-native cloud infrastructure and early #MI355X deployments. https://lnkd.in/gxxxC5SU

From day one, we’ve focused on building AI infrastructure around AMD Instinct™ GPUs and ROCm™ because we believe the future of AI won’t run on a single architecture, vendor, or cloud. Open ecosystems create better outcomes for developers, enterprises, and the future of AI itself. Excited to keep pushing AI advancements forward together and proving what’s possible with open AMD infrastructure.

AI builders in Vegas, come hang out with us. We’re hosting this month’s Las Vegas AI Builders meetup at TensorWave HQ with StartUp Vegas. If you’re working on models, infra, agents, GPUs, startups, or anything in the AI stack, you’ll probably find your people here. No presentations. No corporate conference energy. Register here: https://luma.com/b7zq7vmr

At Web Summit Vancouver, TensorWave CEO Darrick Horton 🌊 joined Rodrigo Liang (SambaNova) and Ami Badani (Arm) to discuss the biggest shifts shaping the future of AI infrastructure; from power constraints and inference to open ecosystems and sovereign AI. Watch the panel discussion here: https://lnkd.in/guYFT_Sj