Acceleration Built For Smarter Spend
The fastest path to compute can become the hardest system to scale. TensorWave helps AI teams move beyond default infrastructure with AMD-powered cloud built for performance, reliability, and long-term control.
ROCm & Roll Karaoke Advancing AI Afterparty
Join TensorWave after AMD’s Advancing AI for karaoke, light bites, and an open bar. Come unwind with the teams building what’s next, grab a mic, and end the day on a higher note.
Thursday, July 23
9:00 PM - 12:00 AM PST
Register to See Address
San Francisco, California
Events
AMD Advancing AI 2026
We’re joining the builders, partners, and infrastructure leaders shaping the next era of AI to showcase what responsible acceleration looks like in practice. AMD-powered infrastructure, open-model architecture, production AI conversations, and the systems work behind it all. Catch TensorWave leadership, live conversations from the show floor, and connect with the community after hours. Here’s what to expect from TensorWave at Advancing AI. July 22-23 | San Francisco, CA | Booth 935 Come for the giveaways. Stay for the conversations!
Powering open-model inference at scale
TensorWave and Featherless AI break down what it takes to serve hundreds of open-weight models at production scale on AMD Instinct™ accelerators. The session brings real performance and cost data to the responsible acceleration conversation: faster systems, better economics, and more control over where AI runs.
Afterparty with TensorWave
Stop by the TensorWave booth to grab your wristband for our karaoke afterparty. Join us after Advancing AI for light bites, an open bar, and open mic karaoke with the builders, partners, and teams pushing AI infrastructure forward.
Meet the team accelerating responsibly
Connect with TensorWave’s team of leaders on the show floor. Whether you’re evaluating AMD compute, planning infrastructure for production AI, or trying to understand what comes next for open-model workloads, our team will be there to talk through the practical decisions behind building responsibly.
Beyond Frontier: Live
The frontier is more than models. It is power, silicon, cooling, networking, software, and the systems work that makes intelligence run. We’ll be recording a new episode of the Beyond Frontier podcast, where TensorWave’s Jeff Tatarchuk and Ilya Tabakh meet the people building the full horizon of AI infrastructure.
Taking the floor
Darrick Horton, Co-Founder and CEO of TensorWave, will join TheCUBE for a show floor interview on the future of AI infrastructure, the growing role of AMD, and what it takes to scale AI without defaulting into closed-ecosystem compromises. It’s a major moment for TensorWave’s point of view at Advancing AI.
Product
AMD Instinct™ MI355X for Scale Without Surrender
TensorWave’s AMD Instinct™ MI355X infrastructure gives AI teams the memory, density, and flexibility to move demanding workloads from experimentation to production. For teams building beyond default infrastructure, MI355X creates a stronger foundation for high-performance training, inference, and open-model scale on AMD.
Performance
288GB HBM3E and high-bandwidth memory help support frontier model development, training, and large-scale inference.
Density
8x MI355X accelerators per standard node deliver concentrated compute for demanding AI infrastructure needs.
Control
Bare metal access, optional Kubernetes, and Slurm support give teams more flexibility in how workloads are deployed and operated.
Reliability
Direct liquid cooling supports sustained performance, improved efficiency, and better thermal stability.
CORPORATE RESPONSIBILITY
Building Infrastructure Also Means Building Stronger Communities
As we grow, we're committed to creating economic opportunity, supporting the open AI ecosystem, investing in local talent, and helping advance the technology community. Our goal is to make a positive impact not only through the infrastructure we provide, but through the people, partnerships, and communities we support.
Growing the Las Vegas Tech Ecosystem
Creating high-quality jobs and investing in the growth of Southern Nevada as a destination for AI and technology innovation.
Supporting Open AI Innovation
Contributing to an open AI ecosystem by championing open-source software, open standards, and developer accessibility.
Investing in Education & Talent
Partnering with universities, students, and community organizations to help develop the next generation of AI engineers and technology leaders.
Building Sustainable Infrastructure
Designing efficient AI infrastructure that helps organizations maximize performance while making responsible use of compute resources.
Strengthening Our Communities
Supporting local initiatives, hosting educational events, and creating opportunities for collaboration across the AI ecosystem.
FAQ
Meet with our team to learn how organizations are using AMD Instinct GPUs for AI training, fine-tuning, and inference. We'll be happy to answer questions, walk through real-world deployments, and discuss your AI infrastructure needs.
Visit us at Booth #935 during the Expo to meet the team, see what we're building, and grab some TensorWave swag. Planning to join us for the afterparty? Be sure to stop by our booth to pick up your wristband.
Yes. Book a meeting with our team before the event to guarantee dedicated time to discuss your AI infrastructure goals.
1. From Open Telco to What's Next: Inside ScalarLM x TensorWave
Hear from Greg Diamos, architect of ScalarLM, as he explores how open infrastructure is making domain-specific AI models a reality. Learn what the team discovered while building ScalarLM on AMD, what's next for the project, and why open AI stacks are shaping the future of enterprise AI.
July 22 | 2:30–2:55 PM PDT | Meet the Experts
2. Powering Open-Model Inference at Scale with TensorWave and Featherless AI
Join TensorWave Founder & CEO Darrick Horton and Featherless AI CEO Eugene Cheah as they discuss how TensorWave built an AMD-first AI cloud on AMD Instinct GPUs and ROCm, along with practical lessons from serving hundreds of open models in production.
July 23 | 1:00–1:45 PM | Room 2012
Yes. Join us for Advancing the After Party: TensorWave Unplugged, an evening of networking, drinks, and karaoke with the AI community. Co-hosted by Trac3 in partnership with Wafer, this event is designed for developers, operators, founders, and anyone passionate about the future of AI. RSVP: https://luma.com/gm18d420