AI workstation
Serve a bigger model without cutting it down first.
48 GB of GDDR7 per GPU changes what fits: longer context, larger batches and higher-precision weights, with the Soika Enterprise licence and A2A network option included.
- 48 GB
- GPU memory
- 1,344 GB/s
- Bandwidth
- 14,080
- CUDA cores
- PCIe Gen 5
- Interface
Key features
48 GB per GPU for larger models and longer context.
- 48 GB of GDDR7 memory per GPU
- 1,344 GB/s of memory bandwidth
- 14,080 CUDA cores, 384-bit memory interface
- Agent-to-Agent network option included
Full specification
- Model
- SOIKASMRTXPROB5000
- Device type
- Multi-GPU workstation, Xeon 60-core platform
- Graphics
- 2× NVIDIA RTX PRO 5000 Blackwell
- Processor
- Intel® Xeon® w9-3595X — 60 cores / 120 threads
- Memory
- 8× 64GB DDR5-4800 2Rx4 ECC RDIMM (512GB)
- Storage
- 8TB NVMe
- GPU memory
- 48 GB GDDR7 with ECC
- Memory interface
- 384-bit
- Memory bandwidth
- 1,344 GB/s
- Network
- 2× 10GbE RJ45 + 1× management LAN
Included with every system
Software, licensed and pre-installed
Soika Stack
GPU clustering, LLM deployment and inference management, pre-configured and licensed.
Mockingjay Network
No-code AI agent design and agent-fleet management with the Enterprise licence.
Ubuntu LTS
Hardened Linux base image with the full CUDA and container toolchain already in place.
- Parts
- 3 years
- Live service
- 3 years, engineer-led
- Support tier
- Standard 3-year; on-site service and SLA available through partners
Related
Other configurations
Sizing
Configure the SM RTX PRO 5000
Tell us your workload and concurrency. We will confirm the configuration, lead time and the partner who delivers it in your region.