Skip to main content

Zyphra: Large-Model Training Lessons on AMD

abstract background

Learn what it took to train ZAYA1-74B, a 74B-parameter mixture-of-experts model, end-to-end on AMD Instinct MI300X. This session shares key engineering lessons from designing an efficient training stack, optimizing long-context performance, and building a reinforcement learning pipeline for math, code, and agentic AI workloads. Discover practical insights for training and deploying large AI models on AMD infrastructure.

July 23, 2026 4:00 PM - 4:15 PM PDT

Beren Millidge
Beren Millidge

Co-Founder & Chief Scientist | Zyphra

Rob Washbourne
Rob Washbourne

Post Training Lead | Zyphra

Rishi Iyer
Rishi Iyer

Member Technical Staff | Zyphra

Topic

Agentic & Generative AI

AI Training & Inference

Session Type

Theater Session

Related Product

Instinct

Related Sessions

View All Sessions
  • Unlock the Power of AI with AI Producer Studio

    AI Producer Studio is optimized for AMD Ryzen AI processors to automate multi-camera meetings, livestreams, and recordings. See how AI detects presenters, follows conversations, and automatically manages camera switching, framing, and layouts in real time. Learn how organizations can simplify professional video production and enhance hybrid collaboration with AMD.;AI Producer Studio is optimized for AMD Ryzen AI processors to automate multi-camera meetings, livestreams, and recordings. See how AI detects presenters, follows conversations, and automatically manages camera switching, framing, and layouts in real time. Learn how organizations can simplify professional video production and enhance hybrid collaboration with AMD.

    July 23, 2026

  • Domain-Specific AI at Scale: Open Models, Post-Training, and AI Infrastructure

    Learn how domain-specific AI moves beyond generic models using post-training, domain evals, and scalable open infrastructure. Using Open Telco Models as a case study, this session covers curated data, reward loops, unified training and serving, and AMD Instinct/ROCm-based stacks for building specialized AI systems at enterprise scale.;Learn how domain-specific AI moves beyond generic models using post-training, domain evals, and scalable open infrastructure. Using Open Telco Models as a case study, this session covers curated data, reward loops, unified training and serving, and AMD Instinct/ROCm-based stacks for building specialized AI systems at enterprise scale.

    July 23, 2026

  • From Models to Production—A Blueprint for AI at Scale

    Moving AI from training to production takes more than GPUs. Hear how Microsoft and Chai AI built scalable AI infrastructure on Vultr using AMD Instinct GPUs and ROCm. Learn best practices for data locality, secure networking, Kubernetes orchestration, benchmarking, cost optimization, and scale-out operations. Leave with a practical blueprint for deploying fast, portable, production-ready AI workloads.;Moving AI from training to production takes more than GPUs. Hear how Microsoft and Chai AI built scalable AI infrastructure on Vultr using AMD Instinct GPUs and ROCm. Learn best practices for data locality, secure networking, Kubernetes orchestration, benchmarking, cost optimization, and scale-out operations. Leave with a practical blueprint for deploying fast, portable, production-ready AI workloads.

    July 23, 2026

  • Training at Scale with AMD Primus

    Primus makes large-scale training on Instinct reliable, debuggable and highly performant. It supports the latest OSS training frameworks, models, and is expanding support to new, cutting-edge model architectures, training techniques, and datatypes. SOTA pre and post training performance with Primus, proven at scales of thousands of GPUs, positions an AMD Instinct GPU as a competitive solution for model development at frontier labs, enterprises, and AI startups.;Primus makes large-scale training on Instinct reliable, debuggable and highly performant. It supports the latest OSS training frameworks, models, and is expanding support to new, cutting-edge model architectures, training techniques, and datatypes. SOTA pre and post training performance with Primus, proven at scales of thousands of GPUs, positions an AMD Instinct GPU as a competitive solution for model development at frontier labs, enterprises, and AI startups.

    July 23, 2026

  • From Tokens to Outcomes: Driving AI ROI with Lenovo Hybrid Infrastructure

    AI success is increasingly measured by business outcomes, not model size. As agentic AI accelerates inference demand, organizations must improve token efficiency, infrastructure utilization, and energy consumption to maximize ROI. Learn how Lenovo Hybrid AI Factories, powered by AMD, help enterprises deploy AI from personal systems to rack-scale environments while reducing token costs, increasing control and utilization, and supporting more sustainable AI growth.;AI success is increasingly measured by business outcomes, not model size. As agentic AI accelerates inference demand, organizations must improve token efficiency, infrastructure utilization, and energy consumption to maximize ROI. Learn how Lenovo Hybrid AI Factories, powered by AMD, help enterprises deploy AI from personal systems to rack-scale environments while reducing token costs, increasing control and utilization, and supporting more sustainable AI growth.

    July 23, 2026

  • HP & AMD Co-engineering Local Compute for Agentic AI

    As AI workloads explode, a cloud-only approach creates friction in data movement, latency, cost, and security. This session explores how keeping compute close to data unlocks measurable advantages for AI workflows. Learn how HP and AMD are co-engineering AI workstations with high-performance local compute and the memory capacity to support larger models and datasets. The future is hybrid—running AI in the right place, at the right time, across environments.;As AI workloads explode, a cloud-only approach creates friction in data movement, latency, cost, and security. This session explores how keeping compute close to data unlocks measurable advantages for AI workflows. Learn how HP and AMD are co-engineering AI workstations with high-performance local compute and the memory capacity to support larger models and datasets. The future is hybrid—running AI in the right place, at the right time, across environments.

    July 23, 2026

  • Agentic AI Needs Open Infrastructure: Here's How to Build It

    The agentic AI stack has evolved to fast multi-model orchestration, tool-augmented reasoning, and long-running inference chains. The hardware conversation hasn't kept up, and many teams default to one GPU vendor without evaluating alternatives. This interactive session is for builders to learn what they're missing. We'll review head-to-head benchmark data from third-party testing, discuss production-ready serving stacks on ROCm, and break down TCO for teams running multi-step agents at scale.;The agentic AI stack has evolved to fast multi-model orchestration, tool-augmented reasoning, and long-running inference chains. The hardware conversation hasn't kept up, and many teams default to one GPU vendor without evaluating alternatives. This interactive session is for builders to learn what they're missing. We'll review head-to-head benchmark data from third-party testing, discuss production-ready serving stacks on ROCm, and break down TCO for teams running multi-step agents at scale.

    July 23, 2026

  • Agentic Kernel Performance Tuning with AMD ROCm

    This session introduces an agentic kernel development workflow for optimizing AI and HPC workloads on AMD ROCm. Learn how a self-directing optimization loop can profile, analyze, optimize, validate, and generate production-ready kernel improvements with minimal manual tuning. The talk highlights how AMD is accelerating kernel engineering by reducing weeks of performance optimization effort into an automated, scalable workflow for developers and performance engineers.;This session introduces an agentic kernel development workflow for optimizing AI and HPC workloads on AMD ROCm. Learn how a self-directing optimization loop can profile, analyze, optimize, validate, and generate production-ready kernel improvements with minimal manual tuning. The talk highlights how AMD is accelerating kernel engineering by reducing weeks of performance optimization effort into an automated, scalable workflow for developers and performance engineers.

    July 23, 2026