B300
NVIDIA · Blackwell Ultra
NVIDIA HGX B300: only 9 GPUs left at $4,899/mo, 25% below market.Only 9 B300 left at $4,899/mo
Configure B300NVIDIA · Blackwell · PCIe 5.0 · up to 8 per server
The NVIDIA RTX PRO 6000 Blackwell Server Edition brings 96 GB of GDDR7 and FP4 Tensor Cores to a PCIe card. It is the cheapest way to serve models up to about 70B, and the best GPU in our line-up for rendering, video and image generation.
PCIe cards are not NVLink-connected: use them for workloads that scale per GPU or with pipeline parallelism.
One flat price per GPU, the same from 1 to 8 GPUs. The market median is the public on-demand list price of specialist GPU clouds × 730 hours.
| Server | CoreGPU / month | Market median / month | You save / month | You save / year | ≈ per GPU-hour |
|---|---|---|---|---|---|
| 1× RTX PRO 6000 | $1,199 | $400 | $4,800 | $1.64 | |
| 2× RTX PRO 6000 | $2,398 | $800 | $9,600 | $1.64 | |
| 4× RTX PRO 6000 | $4,796 | $1,600 | $19,200 | $1.64 | |
| 8× RTX PRO 6000 | $9,592 | $3,200 | $38,400 | $1.64 |
Prices in USD, excluding taxes. Market median checked Nov 4, 2026 across specialist GPU clouds; hyperscalers excluded. All GPU prices
RTX PRO 6000 has twice the memory (96 vs 48 GB) and Blackwell FP4; L40S is cheaper for small models and classic rendering.
| RTX PRO 6000 | L40S | |
|---|---|---|
| Architecture | Blackwell | Ada Lovelace |
| GPU memory | 96 GB GDDR7 | 48 GB GDDR6 |
| Memory bandwidth | 1.6 TB/s | 864 GB/s |
| FP8 dense | 1 PF | 0.73 PF |
| FP4 dense | 2 PF | n/a |
| Interconnect | PCIe 5.0 x16 | PCIe 4.0 x16 |
| Power (TDP) | 600 W | 350 W |
| Price per GPU / month | $1,199 | $849 |
| Price per GB of memory | $12.49 | $17.69 |
| Price per FP8 PFLOPS | $1,199 | $1,163 |
| Availability | 52 available | 38 available |
Per-GPU figures from the vendor datasheet. Tensor throughput is dense (without sparsity).
| Model | NVIDIA RTX PRO 6000 Blackwell Server Edition |
|---|---|
| Architecture | Blackwell |
| Form factor | PCIe 5.0 · up to 8 per server |
| GPU memory | 96 GB GDDR7 |
| Memory bandwidth | 1.6 TB/s |
| FP8 Tensor (dense) | 1 PF |
| FP4 Tensor (dense) | 2 PF |
| Interconnect | PCIe 5.0 x16 |
| Power (TDP) | 600 W |
| GPUs per server | 1 to 8 |
| Availability | 52 available |
Another question? Our engineers answer within one business day.
Contact us$1,199 per GPU per month at CoreGPU, about $1.64 per GPU-hour, versus a market median of $1,599 ($2.19/hour × 730 h). A full 8-GPU server is $9,592 per month, taxes excluded.
Yes: 52 available in the current allocation of 96 GPUs. In-stock servers are delivered with root SSH access in under 60 seconds after payment.
From 1 to 8 GPUs per server, at the same price per GPU. These are PCIe cards, so they scale per GPU rather than over NVLink. Need more? Rent several servers or ask for a private allocation.
RTX PRO 6000 has twice the memory (96 vs 48 GB) and Blackwell FP4; L40S is cheaper for small models and classic rendering. See the full RTX PRO 6000 vs L40S comparison.
Dedicated GPUs (never shared or sliced), root SSH access, your choice of a clean OS or a ready image with CUDA, PyTorch and vLLM, local NVMe scratch, and month-to-month billing with no setup fee.
NVIDIA · Blackwell Ultra
AMD · CDNA 4
AMD · CDNA 4
Pick 1 to 8 GPUs in the configurator above. Month to month, cancel anytime.