Home / Servers / 8× NVIDIA RTX 5090 32 GB
Canonical hardware · not a commercial offer
8× NVIDIA RTX 5090 32 GB
252 GB system RAM · XEON® GOLD 6530
Hardware
- GPU
- NVIDIA RTX 5090 32 GB
- GPU count
- 8
- VRAM per GPU
- 32.0 GB advertised
- Aggregate VRAM
- 256.0 GB advertised
- System RAM
- 252 GB
- CPU
- XEON® GOLD 6530
- Storage
- 3217.5 GB SSD
- Network
- 2726.1 down · 1394.1 up
AI memory capability
Default catalog below is Q4 at 8K. Statuses reuse Phase 2 (1 GPU) or Phase 8 (2–4 identical GPUs). GPU counts above 4 are not modeled. Public wording is memory-capable, not serving throughput.
Memory-capable at Q4/8K: 0 published models
Serving performance
Measured vLLM serving results for this GPU identity (8× NVIDIA RTX 5090 32 GB), not a specific rental host. Tensor parallel size is stored explicitly and is not assumed equal to GPU count. Request rate is separate from concurrency; this corpus is closed-loop / max-concurrency.
No measured vLLM serving rows map to this GPU count yet.
Current rental offers
| Provider | Location | Price | Stock | |
|---|---|---|---|---|
| Vast.ai | CN | $3.202/hour · ~$2338/month equivalent (hourly × 730) | Listed as available | Provider |
Want to own the hardware? Build a PC · Prefer a prebuilt? Workstations