SPARKLE Intel Arc Pro B60 24GB GDDR6 Passive Graphic Card
SPARKLE Intel Arc Pro B60 24GB GDDR6 Passive Graphics Card delivers enterprise-class AI and workstation performance with Intel Xe2 (Battlemage) architecture, 24GB GDDR6 memory, 160 Intel XMX AI Engines, 197 INT8 TOPS, and PCIe Gen5 x8 connectivity. Its fanless passive cooling design makes it an ideal solution for AI inference, local LLM deployment, CAD, rendering, virtualization, engineering, and rack-mounted enterprise server environments.
SPARKLE Intel Arc Pro B60 24GB GDDR6 Passive Graphic Card
Features
- Professional Intel Arc Pro B60 workstation and server GPU.
- 24GB GDDR6 VRAM for AI models, CAD, and professional visualization.
- Built on the latest Intel Xe2 (Battlemage) architecture.
- 160 Intel XMX AI Engines for accelerated AI inference and machine learning.
- Delivers up to 197 INT8 TOPS AI performance.
- High-speed PCIe Gen5 x8 interface.
- Fanless passive cooling optimized for high-airflow rack servers.
- 200W Total Board Power for enterprise-class performance.
- Supports 4 × DisplayPort 2.1 outputs for multi-display workstations.
- Hardware Ray Tracing and DirectX 12 Ultimate support.
- AV1 hardware encode/decode for professional media production.
- Enterprise-certified drivers for AI, CAD, BIM, rendering, engineering, and visualization workloads.
Specifications
| Product Name | SPARKLE Intel Arc Pro B60 24GB GDDR6 Passive Graphic Card |
| Manufacturer | SPARKLE |
| GPU | Intel Arc Pro B60 |
| GPU Architecture | Intel Xe2 (Battlemage) |
| Graphics Memory | 24GB GDDR6 |
| Memory Interface | 192-bit |
| Memory Bandwidth | Up to 456 GB/s |
| AI Engines | 160 Intel XMX AI Engines |
| Xe Cores | 20 |
| Ray Tracing Units | 20 |
| AI Performance | Up to 197 INT8 TOPS |
| Graphics Clock | Up to 2400 MHz |
| PCI Express Interface | PCIe 5.0 x8 |
| Cooling | Fanless Passive Cooling |
| Form Factor | Dual-Slot |
| Total Board Power (TBP) | 200W |
| Display Outputs | 4 × DisplayPort 2.1 |
| DirectX Support | DirectX 12 Ultimate |
| OpenGL Support | OpenGL 4.6 |
| AV1 Hardware Encode/Decode | Yes |
| Target Workloads | AI Inference, LLMs, CAD, 3D Rendering, Engineering, Virtualization, Enterprise Servers |





