Zyphra and AMD Partner to Power Zyphra Cloud on AMD Instinct™ MI355X GPUs

Rhea-AI Summary

Zyphra launched Zyphra Cloud on May 4, 2026, a full-stack AI platform running on AMD Instinct MI355X GPUs via TensorWave infrastructure. Zyphra Cloud debuts with Zyphra Inference, a serverless inference service offering frontier open-weight models (DeepSeek V3.2, Kimi K2.6, GLM 5.1) and long-context, high-throughput inference for agentic, research, and automation workloads. The platform is available today and plans future capabilities including distributed RL/fine-tuning, EPYC CPU sandboxes, and dedicated GPU clusters.

Positive

• Zyphra Cloud launch with Zyphra Inference available today
• Runs on AMD Instinct MI355X GPUs via TensorWave infrastructure
• Supports frontier open-weight models: DeepSeek V3.2, Kimi K2.6, GLM 5.1
• Serverless inference optimized for long-context, high-throughput agentic workloads
• Roadmap includes distributed RL/fine-tuning and dedicated GPU clusters

Negative

• None.

News Market Reaction – AMD

-5.27%

160 alerts

-5.27% Session close to close

+19.7% Peak Tracked

-2.5% Trough Tracked

$579.19B Market Cap

0.6x Rel. Volume

In the May 4 session, AMD declined 5.27%, reflecting a notable negative market reaction. Argus tracked a peak move of +19.7% during that session. Argus tracked a trough of -2.5% from its starting point during tracking. Our momentum scanner triggered 160 alerts that day, indicating very high trading interest and price volatility.

Data tracked by StockTitan Argus on the day of publication.

Market Context

The stock moved -5.3% in the session following this news. A negative reaction despite a constructive...

Analysis

The stock moved -5.3% in the session following this news. A negative reaction despite a constructive AI partnership would contrast with the prior Meta agreement, which saw a 8.77% gain. Any pullback could reflect profit-taking near the 52-week high at $362.79 or concerns about how quickly partnerships translate into revenue. With short interest about 2.2% and days-to-cover at 1.17, downside pressure would likely stem more from long holders than an aggressive short base.

Key Figures

DeepSeek model version: V3.2 Kimi model version: K2.6 GLM model version: 5.1

3 metrics

DeepSeek model version V3.2 DeepSeek V3.2 offered via Zyphra Inference

Kimi model version K2.6 Kimi K2.6 offered via Zyphra Inference

GLM model version 5.1 GLM 5.1 offered via Zyphra Inference

Previous Partnership Reports

1 past event · Latest: Feb 24 (Positive)

Same Type Pattern 1 events

Date	Event	Sentiment	24h Move	Catalyst
Feb 24	AI partnership deal	Positive	+8.8%	Multi‑year Meta partnership deploying up to 6 GW of AMD Instinct GPUs.

24h Move is the share-price change in the day after each event; other market factors may also have contributed.

Pattern Detected

The only recent partnership headline showed a strong positive price reaction, indicating investors have rewarded large AI infrastructure collaborations.

Recent Company History

Recent AMD news has centered on large-scale AI and data center partnerships and events. A prior Feb 24, 2026 partnership with Meta to deploy up to 6 gigawatts of AMD Instinct GPUs drove a 8.77% one-day gain, tied to expectations for multi‑year revenue growth and EPS accretion. Additional AI-related announcements and events highlight AMD’s focus on its Instinct GPU and EPYC CPU ecosystem, which aligns with today’s Zyphra Cloud partnership using AMD Instinct MI355X GPUs.

Key Terms

serverless inference, open-weight models, reinforcement learning, fine-tuning, +2 more

6 terms

serverless inference technical

"Zyphra Cloud launches with Zyphra Inference, a serverless inference service providing"

Serverless inference is a way to run artificial intelligence models on demand without a company owning or managing the underlying servers; the cloud provider automatically supplies computing power when a request comes in and bills only for actual usage. For investors, it matters because it lets businesses add AI features quickly, scale up or down with customer demand, and convert large upfront infrastructure costs into smaller, predictable operating expenses — which can improve margins and speed product rollout.

open-weight models technical

"service providing access to frontier open-weight models including DeepSeek V3.2, Kimi"

Open-weight models are machine-learning systems whose internal parameters (the numeric “weights” that determine how inputs are turned into outputs) are publicly available for inspection, modification, and reuse. For investors, that transparency makes it easier to audit performance claims, assess risks and biases, and accelerate product development—think of it as a recipe where every ingredient and measurement is shown, so anyone can test, tweak, or reproduce the result.

reinforcement learning technical

"post-training services such as reinforcement learning and fine-tuning, sandboxed agent"

A type of artificial intelligence that learns by trial and error, receiving feedback from its actions to favor choices that lead to better outcomes. Think of it like a salesperson learning which pitches close deals by trying different approaches and keeping the ones that work. For investors, reinforcement learning matters because it can power smarter trading systems, optimize business operations, or improve products—potentially boosting efficiency and profits while also introducing model and execution risks.

fine-tuning technical

"post-training services such as reinforcement learning and fine-tuning, sandboxed agent"

Fine-tuning means making small, deliberate adjustments to a company’s tools, models, processes or plans to improve performance or accuracy without overhauling the whole system. Like tightening the strings on a guitar for a clearer note, these tweaks can reduce risk, cut costs, boost efficiency or sharpen forecasts, so investors watch fine‑tuning as a signal management is optimizing resources and responding to market or regulatory changes.

bare-metal infrastructure technical

"access to dedicated GPU clusters and bare-metal infrastructure. Together, these"

Physical servers and networking equipment leased or owned and dedicated to a single user or workload, run without shared hosting layers or multi-tenant software. For investors, bare-metal infrastructure matters because it often delivers faster, more predictable performance, stronger security controls and clearer cost trade-offs—similar to owning a car instead of riding a shared taxi—affecting a company’s operating costs, scalability and competitive positioning.

gpu technical

"Powered by AMD Instinct™ MI355X GPUs on TensorWave's purpose-built infrastructure,"

A GPU (graphics processing unit) is a specialized computer chip designed to handle many calculations at once, originally for rendering images and video but now widely used for tasks like artificial intelligence, data analysis and high-performance computing. Investors watch GPU demand and prices because strong sales often signal growth for chip makers and their customers, affect profit margins and capital spending, and can forecast wider trends in gaming, AI adoption and cloud services.

AI-generated analysis. How Rhea-AI works. Not financial advice.

See more from StockTitan in Google Search and AI answers. Adds StockTitan as a preferred source · opens Google

Add on Google Not now

Zyphra announced Zyphra Cloud, a full-stack AI platform on AMD powered by Tensorwave. The platform launches with Zyphra Inference, a serverless inference service for frontier open-weight models focused on long-horizon agentic workloads.

SAN FRANCISCO, May 4, 2026 /PRNewswire/ -- Zyphra today announced Zyphra Cloud, a full-stack AI platform that brings advanced innovations from Zyphra Research into production for developers, enterprises, and frontier AI hyperscalers. Powered by AMD Instinct™ MI355X GPUs on TensorWave's purpose-built infrastructure, Zyphra Cloud unifies model serving, agent infrastructure, and scalable compute into a single platform for building and deploying advanced AI systems.

Zyphra Cloud launches with Zyphra Inference, a serverless inference service providing access to frontier open-weight models including DeepSeek V3.2, Kimi K2.6, and GLM 5.1. Zyphra Inference combines custom kernels, novel long-context inference algorithms, and advanced parallelism schemes to deliver high-throughput, low-latency performance for production-grade long-horizon use cases such as agentic coding, deep research, and long-horizon workflow automation.

"Zyphra Cloud is the natural extension of our research. We've spent years building, optimizing, and validating AI systems on AMD infrastructure, and are now bringing that capability to market as a platform for developers and enterprises, starting with production-grade inference through Zyphra Inference," said Krithik Puthalath, Founder and CEO of Zyphra. "With Zyphra Cloud, teams can build and deploy advanced AI systems on AMD with the performance, efficiency, and scale required for real-world workloads."

"AMD delivers leadership solutions, in combination with open platforms and deep industry-wide collaborations, to power the next generation of AI infrastructure," said Negin Oliver, Corporate Vice President, Business Development for AI, AMD. "Zyphra Inference running on AMD Instinct MI355X GPUs, on TensorWave compute infrastructure, demonstrates how optimized AI software combined with our accelerator architecture can deliver leading AI inference performance in production environments for the most demanding open-weight models available today."

"TensorWave exists to give AI-native companies like Zyphra the dedicated, high-performance AMD compute they need without compromise," said Jeff Tatarchuk, Co-Founder and Chief Growth Officer of TensorWave. "Powering Zyphra Inference with our MI355X infrastructure is exactly the kind of partnership we built TensorWave for — enabling teams to ship production-ready AI on the latest AMD accelerators at scale."

Expanding the Platform

Zyphra Cloud is designed to expand beyond inference into a broader integrated platform. Upcoming capabilities include distributed post-training services such as reinforcement learning and fine-tuning, sandboxed agent and development environments powered by AMD EPYC™ CPUs, and access to dedicated GPU clusters and bare-metal infrastructure. Together, these components provide a unified environment for building, training, and deploying AI systems.

Availability

Zyphra Cloud is available today. For more information visit www.zyphra.com/cloud or to access Zyphra Cloud directly go to cloud.zyphra.com.

About Zyphra

Zyphra is an open superintelligence research and product company based in San Francisco, CA, on a mission to build human-aligned AI that helps individuals and organizations reach their fullest potential. For more information visit www.zyphra.com or contact cloud@zyphra.com.

About TensorWave

TensorWave is the AI and HPC cloud purpose-built for performance. Powered exclusively by AMD Instinct™ Series GPUs, TensorWave delivers high-bandwidth, memory-optimized infrastructure that scales with the most demanding AI workloads. Backed by AMD Ventures and Magnetar, TensorWave is among the first cloud providers to deploy AMD Instinct™ MI355X GPUs. For more information visit tensorwave.com.

About AMD

AMD (NASDAQ: AMD) is a global semiconductor leader powering the products and services that help solve the world's most important challenges. For more information, visit www.amd.com.

Media Contact:
Paul White
Chief Business Officer
press@zyphra.com
www.zyphra.com

View original content to download multimedia:https://www.prnewswire.com/news-releases/zyphra-and-amd-partner-to-power-zyphra-cloud-on-amd-instinct-mi355x-gpus-302761765.html

SOURCE Zyphra

FAQ

What is Zyphra Cloud and how is it powered (AMD ticker AMD)?

Zyphra Cloud is a full-stack AI platform combining model serving and agent infrastructure. According to the company, it runs on AMD Instinct MI355X GPUs using TensorWave compute infrastructure and is designed for production-grade long-horizon AI workloads.

What does Zyphra Inference offer for developers and enterprises (AMD)?

Zyphra Inference is a serverless inference service providing access to frontier open-weight models. According to the company, it delivers custom kernels, long-context algorithms, and parallelism for high-throughput, low-latency production inference.

Which models are available on Zyphra Cloud at launch (AMD)?

At launch Zyphra Cloud supports DeepSeek V3.2, Kimi K2.6, and GLM 5.1. According to the company, these frontier open-weight models are accessible through Zyphra Inference for long-horizon workloads.

Is Zyphra Cloud available now and how can I access it (AMD)?

Yes, Zyphra Cloud is available today for developers and enterprises. According to the company, users can learn more or access the platform via the Zyphra Cloud website and direct cloud.zyphra.com links.

What future capabilities will Zyphra Cloud add and when (AMD)?

Zyphra Cloud plans to add distributed post-training services and sandboxed EPYC CPU environments. According to the company, upcoming features include reinforcement learning, fine-tuning, bare-metal GPU clusters, and dedicated infrastructure support.