Lenovo unveils new AI inferencing servers to accelerate real-time enterprise workloads
Lenovo unveils new AI inferencing servers to accelerate real-time enterprise workloads
Lenovo has expanded its enterprise AI portfolio with a new lineup of inferencing-focused servers, solutions, and services unveiled during Tech World @ CES 2026 at Sphere in Las Vegas.
The announcement strengthens Lenovo’s Hybrid AI Advantage, with new infrastructure designed to help organizations deploy real-time AI workloads across cloud, data center, and edge environments. Lenovo says the move addresses a shift from training large AI models to running inference at scale in real-world applications.
AI inferencing plays a critical role in turning trained models into actionable intelligence. It enables instant decision-making in environments such as retail, healthcare, manufacturing, and financial services. Market research firm Futurum estimates the global AI inference infrastructure market will grow from USD 5.0 billion in 2024 to USD 48.8 billion by 2030.
“Enterprises need AI that delivers insight the moment data is created,” said Ashley Gorakhpurwalla, Executive Vice President at Lenovo and President of its Infrastructure Solutions Group. “Our inferencing-optimized infrastructure helps customers turn data into real-time intelligence that drives faster decisions and innovation.”
New ThinkSystem and ThinkEdge servers power AI inference at scale
Lenovo introduced new ThinkSystem and ThinkEdge servers built specifically for AI inferencing workloads of different sizes and environments.
The ThinkSystem SR675i targets large-scale deployments that require running full large language models. This supports use cases in manufacturing simulations, healthcare diagnostics, and financial modeling. The ThinkSystem SR650i focuses on accelerated inferencing in existing data centers, offering high-density GPU performance with easier deployment.
For edge scenarios, Lenovo unveiled the ThinkEdge SE455i. The compact server delivers ultra-low latency AI inference for retail, telecom, and industrial environments where data is generated. Lenovo says the system operates reliably in extreme conditions.
Across the portfolio, Lenovo integrates its Neptune air and liquid cooling technologies to address energy efficiency challenges. The servers are also available through Lenovo TruScale, allowing enterprises to adopt AI infrastructure through a pay-as-you-go model.
Hybrid AI platforms target faster deployment and lower risk
The new servers form the hardware foundation of the Lenovo Hybrid AI Factory. It is a modular framework designed to simplify enterprise AI deployment.
Lenovo says the platform unifies compute, storage, networking, and orchestration into a single system, helping organizations reduce deployment time and operational risk. The Hybrid AI Factory supports multiple software ecosystems, including Nutanix AI, Red Hat AI, and Canonical Ubuntu Pro, The AI Factory aimed to help enterprises flexibility based on workload complexity and scale.
These platforms support agentic AI and shared inference environments, enabling organizations to maximize GPU utilization while maintaining security and performance.
New services aim to accelerate AI adoption
Alongside new infrastructure, Lenovo introduced Hybrid AI Factory Services focused on AI inferencing. The services provide advisory, deployment, and managed support to help enterprises design and operate inferencing environments tailored to specific industries.
Lenovo says the services help organizations achieve faster returns on investment, backed by Lenovo Premier Support and flexible TruScale financing options as AI operations scale.