PLATINUM-96 — professional workstation.
The design we run our own work on, one revision on. One 96 GB card with ECC, which is the point: it holds a 70B model at 4-bit with a long context, and ECC earns its keep once a training run lasts days.
Best for QLoRA fine-tuning past 100B at 4-bit, full-weight fine-tuning to roughly 8B on one GPU, production inference with vLLM continuous batching, molecular dynamics, diffusion training, and multi-user JupyterHub. 96 GB also holds a 70B model at 4-bit with a long context. ECC memory earns its keep once a run lasts days. PLATINUM-96 is the design we run our own work on, one revision on.
Bill of materials
| GPU | 1× RTX PRO 6000 Blackwell Workstation Edition — 96 GB GDDR7 ECC, 1,792 GB/s, 512-bit, PCIe 5.0 ×16, 600 W maximum power |
|---|---|
| CPU | AMD Ryzen 9 9950X — 16C/32T, Zen 5, TSMC 4 nm, 80 MB cache, 4.3 GHz base / 5.7 GHz boost, 170 W TDP |
| Memory | 128 GB DDR5-6000 CL36 (2×64 GB), EXPO — two DIMMs rather than four, because four dual-rank sticks will not hold 6000 on AM5 |
| Storage | 2 TB PCIe 5.0 NVMe (Crucial T700) on the CPU-attached M.2 + 8 TB PCIe 4.0 NVMe for datasets. AM5 does not have the lanes for a second Gen5 drive plus U.2 alongside a ×16 GPU — that configuration needs TRX50. |
| Power | Seasonic Vertex PX-1200 — 1,200 W 80+ Platinum, ATX 3.1, native 12V-2×6. ~940 W under load, so about 78% of rating. |
| Motherboard | ASUS ProArt X870E-Creator WiFi, or an equivalent AM5 board with PCIe 5.0 ×16 plus a Gen5 M.2 |
| Cooling & chassis | 360 mm AIO on the CPU; full-tower with clearance for a 304 mm card |
| Software | Ubuntu 24.04 LTS · CUDA 12.8 · Docker 27.x · k3s · MLflow |
Three starting points, each sized against a model it has to hold. If none of them is the right fit we adjust the spec to what you actually plan to run.
Get a build quoted.
Tell us the model, the data and the constraint. We will come back with a spec and an itemised price.
Start a project