Agentic AI is changing data center workloads. Agents read far more than they write: long contexts, multistep workflows, compute dominated by prefill. Intel D-series GPUs are built for this new era of agentic AI, efficiently handling larger contexts, more concurrent sessions, and more tokens within practical power and cost limits.
Intel GPU D270 highlights
- Memory up to 480 GB LPDDR5X per card
- 350 W air-cooled, passive cooling in standard racks
- 32 Xᵉ3P cores, 256 XMX matrix engines
- Standard PCIe 5.0 form factor; no proprietary networking required
- Die area reallocated from graphics and ray-tracing to inference-focused Xᵉ3P cores
- Open software positioning: vLLM, PyTorch, SGLang, OpenAI Triton (where approved)
Intel positions D270 for agentic fleets, whole-context reasoning, and efficient token serving without HBM, targeting enterprise AI, SaaS, GPUaaS, and sovereign-cloud deployments.
Performance projections vs RTX Pro 6000 are based on publicly available information and Intel internal analysis as of Feb 2026; disclaimers note performance varies by use and configuration.