NVIDIA · Blackwell
Nosana network, hosts worldwide
Stock and price checked 16 min ago.
Self-service: you start it yourself, with no order or sales contact.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
nvidia-pro6000-6x
Nosana network, hosts worldwide
Stock and price checked 16 min ago.
Self-service: you start it yourself, with no order or sales contact.
Confidential computing: data stays encrypted while the GPU works on it, so the host can't read it.
nvidia-pro-6000-cc
Canada · 224 vCPU · 1800 GB RAM · 100GB root + 27600GB ephemeral
Stock and price checked 16 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-RTX-PRO6000-SEx8
Canada · 28 vCPU · 225 GB RAM · 100GB root + 3300GB ephemeral
Stock and price checked 16 min ago.
You start it yourself once there is stock.
n3-RTX-PRO6000-SEx1
Canada · 56 vCPU · 450 GB RAM · 100GB root + 6800GB ephemeral
Stock and price checked 16 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-RTX-PRO6000-SEx2
Canada · 112 vCPU · 900 GB RAM · 100GB root + 13800GB ephemeral
Stock and price checked 16 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
n3-RTX-PRO6000-SEx4
Austria · 248 vCPU · 960 GB RAM · Storage billed separately
Stock and price checked 16 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
gpurtx6000pro.huge
Switzerland · 248 vCPU · 960 GB RAM · Storage billed separately
Stock and price checked 16 min ago.
You start it yourself once there is stock.
This GPU has no fast GPU-to-GPU link. These are separate cards: good for separate jobs, slow for one model split across them.
gpurtx6000pro.huge
Excluding VAT.
Each provider counts once. Full price and spot are separate.
Full price. Offers we can't say are really sold there are left out until they've been in stock once.
Only providers that report a number. Each line is one provider and one server size, because their counts cover different things and can't be added up. Providers that only say yes or no are under Stock by provider.
In stock, yes or no, for every provider selling this GPU. A filled band means in stock: green at full price, orange as spot.
The server sizes you can rent right now.
Can't be linked. This GPU has no NVLink, so several in one server only talk over PCIe at up to 128 GB/s. Each card keeps its own 96 GB. That suits running a separate job on each card. Splitting one model across them works, but slowly.
Headline values are Server Edition (TDP up to 600 W, configurable). Workstation editions list 1792 GB/s memory bandwidth. No NVLink; host link PCIe Gen5. memoryDataRateGbps is derived for Server Edition (1597 GB/s x 8 / 512 bits); Workstation and Max-Q run 28 Gbps. cooling is the passive air-cooled Server Edition; a liquid-cooled variant also exists. processNode, transistorsBillion and dieSizeMm2 are the GB202 figures from NVIDIA's RTX PRO Blackwell whitepaper, which lists the Workstation and Max-Q editions only. pcieBandwidthGBs is PCIe Gen5 x16, 128 GB/s per NVIDIA.