6 providers · per GPU-hour
NVIDIA RTX PRO 6000 cloud pricing
6 offers from 6 providers, 6 with a published on-demand price. Per GPU-hour, US$. Prices checked 28 Sept 2026.
- Lowest on-demand
- $1.80/GPU-hr
- Lowest spot
- $1.39/GPU-hr
- Memory
- 96 GB GDDR7 ECC
Spec sheet: RTX PRO 6000 Blackwell Server Edition ↗
| # | Provider | Offer | On-demand | Spot | Committed | As published (original currency) · source | Actions |
|---|---|---|---|---|---|---|---|
| 1 | Nebius | RTX PRO 6000 x1 | $1.80/GPU-hr | — | — | ||
| 2 | Verda (formerly DataCrunch) | RTX PRO 6000 x1 | $1.84/GPU-hr | — | — | ||
| 3 | Hyperstack | RTX PRO 6000 x1 | $1.85/GPU-hr | — | — | ||
| 4 | Jarvislabs | RTX PRO 6000 x1 1× RTX PRO 6000 Blackwell Server Edition · 28 vCPU · 160 GB RAM | $1.89/GPU-hr | — | — | ||
| 5 | CoreWeave | NVIDIA RTX PRO 6000 Blackwell Server Edition, High Memory (8x) 8× RTX PRO 6000 Blackwell Server Edition · 128 vCPU · 1024 GB RAM | $2.50/GPU-hr | $1.39/GPU-hr | — | ||
| 6 | Cyfuture | RTX PRO 6000 x1 1× RTX PRO 6000 Blackwell Server Edition · 16 vCPU · 128 GB RAM | $2.70/GPU-hr | — | — |
What is the RTX PRO 6000 used for?
The RTX PRO 6000 is a data-centre GPU positioned by its maker for both training and inference, built on the Blackwell architecture and released in 2025.
How large a model fits on one RTX PRO 6000?
With 96 GB of memory, one card holds the weights of a model up to about 48 billion parameters at 16-bit precision, or about 96 billion at 8-bit. That is weights only: inference also needs room for the KV cache, and training needs several times more for gradients and optimiser state, so larger jobs are split across multiple GPUs.
| Spec (as published) | RTX PRO 6000 Blackwell Server Edition |
|---|---|
| Architecture | Blackwell |
| Form factor | PCIe dual-slot FHFL (air) | single-slot FHXL (liquid) |
| Memory | 96 GB GDDR7 ECC |
| Memory bandwidth | 1,597 GB/s |
| Max power (TDP) | 600 W |
| Precisions listed | FP4, FP8, FP16, BF16, TF32, FP32 |
| Partitioning | Up to 4 MIG instances |
| Released | 2025 |
| Source | maker spec sheet ↗ |
Need RTX PRO 6000 capacity for a longer run?
Committed and cluster pricing is often negotiated. Tell us the GPU count, duration and region.