B200 GPU Rental
Rent B200 GPUs on Lium from $5.50 per GPU-hour, built for large-scale AI training, inference, and HPC. Billed per second with no minimum, deployed in seconds from the browser or the CLI.
Price per GPU-hour
$5.50 – $6.40
Live asks across all available pods, billed per second
Available now
4 pods
11 GPUs in total
Reference price
$5.60
Marketplace baseline per GPU-hour
About the B200
The B200 is NVIDIA’s Blackwell data center GPU and the successor to the H100. Two things matter in practice: memory and low precision. Each card carries 180 GB of HBM3e at 8 TB/s, more than twice the capacity and bandwidth of an H100, and Blackwell adds native FP4 alongside FP8, so inference kernels that can drop precision get another step up in throughput. NVLink runs at 1.8 TB/s between cards in a node. The card draws up to 1,000 W, which is why it turns up in purpose-built eight-GPU systems rather than in retrofitted racks.
What people run on it
- Serving 70B+ models on a single card
- FP4 and FP8 inference at scale
- Multi-node training over NVLink
- Long-context and high-batch workloads
B200 specifications
| VRAM | 180 GB HBM3e |
|---|---|
| Memory bandwidth | 8,000 GB/s |
| FP16 tensor | 2.25 PFLOPS |
| FP8 tensor | 4.5 PFLOPS |
| Interconnect | NVLink 1.8 TB/s |
| Power | 1,000 W |
| Architecture | Blackwell |
| Released | 2024 |
Tensor figures are NVIDIA’s published dense numbers for the card, without sparsity; sparsity doubles them. What a pod actually delivers depends on the model, the kernels and the host it runs on.
What runs on one B200
With 180 GB of memory, one card holds roughly the following open models for inference, counting the weights plus about 20% for activations and cache. Rent several GPUs in one pod to go bigger.
| Precision | Largest model that fits |
|---|---|
| FP16 | 70B parameters |
| 8-bit | 70B parameters |
| 4-bit | 70B parameters |
Running one B200 around the clock for a month (730 hours) costs about $4,015 at the lowest live ask; you pay only for the seconds a pod runs.
Frequently asked questions
How much does it cost to rent a B200 on Lium?
B200 pods are listed at $5.50 to $6.40 per GPU-hour right now, across 4 pods with 11 GPUs. Each provider sets its own price and the range moves as pods are added and removed.
Where is the cheapest place to rent a B200 by the hour?
On Lium, B200 pods start at $5.50 per GPU-hour right now, the lowest live ask across 4 pods, and billing is per second with no minimum, so a 20-minute job costs a third of that. Providers set their own prices, so the lowest ask moves as pods are listed and rented.
Is there a minimum rental period?
No. Lium bills per second with no minimum and no reservation: start the pod when you need it, delete it when the job is done, and pay for the seconds it ran.
How do I rent a B200 pod?
Top up a balance by card or crypto, pick a B200 pod on the marketplace, choose a Docker template, add your SSH key and deploy. From a script, run lium up with the node id after exporting LIUM_API_KEY.
Where do the machines run, and what happens to my data?
Providers run their own hardware in their own facilities, and every pod on the marketplace names its city and country, so you can pick a region before you start. Your data lives on the machine you rented and is released when you delete the pod; attach a volume if you need storage that outlives a single pod.
For AI agents — CLI install, agent skill, headless signup: llms-full.txt