NVIDIA Blackwell Ultra
Rent GB300 servers, reserved from data centers
GB300 pairs a Grace CPU with two Blackwell Ultra GPUs, and is mostly sold as the GB300 NVL72 rack: 72 GPUs in one NVLink domain that behaves like a single very large accelerator. It is aimed at training and serving the largest models, where the rack-scale NVLink matters more than the per-GPU price.
- Memory per GPU
- 288 GB HBM3e
- Architecture
- Blackwell Ultra
- Listings
- On request
- From
- Ask
GB300 capacity listed now
No GB300 capacity is listed right now. Send a request and we will source it from our owners.
Tell us what you needWhat GB300 is good for
- Training frontier-scale models across one NVLink domain
- Serving very large mixture-of-experts models with long context
- Reasoning workloads that need many GPUs answering one request
What to ask before you reserve GB300 capacity
- Whether you get a whole NVL72 rack or a slice of one, and how many racks share the network.
- The liquid-cooling and power arrangement, and who is responsible when a tray fails.
- How the servers are interconnected (InfiniBand or RoCE, and at what speed per GPU) if you will train across nodes.
- Whether the price covers storage, egress and support, or only the GPU-hours.
- The uptime they commit to, and what you get back when they miss it.
Questions
- How much does it cost to rent a GB300?
- Owners set their own price per GPU-hour, which depends on the number of servers, the term and the start date. Send a request with what you need and we come back with offers.
- What is the minimum term for reserved GB300 capacity?
- Usually one month or more, set by each owner. Longer terms usually get a lower price per GPU-hour.
- How much memory does a GB300 have?
- 288 GB HBM3e per GPU (NVIDIA Blackwell Ultra).
- Can I rent a single GB300 server?
- Yes, where the owner allows it: each listing shows how many servers it has and you can ask for one or for all of them.
Tell us what you need
Sign in to send a request. It opens a conversation with the owners and the Paralon team, and you get an email for every reply.
Sign in to requestFree, with Google, GitHub or an email code. No commitment.