NVIDIA B200 SXM pricing.
prices as of · 6 providers · 17 configurations
The training board of the Blackwell generation, eight to a node with NVLink between them.
The cheapest B200 today is B200 · Community Cloud at RunPod, $5.98/GPU-hr, read 17 Sept 2026; the cheapest interruptible rate is $3.12/GPU-hr at Verda (DataCrunch).
Cheapest on demand
$5.98
per GPU-hour at RunPod
Median of the spread
$6.69
across 15 configurations
Cheapest interruptible
$3.12
8 of 15 publish one
Per GB of memory
$0.0332/GB-hr
market median $0.0300/GB-hr
What the board is
published by NVIDIA, not by a provider- Memory
- 180 GB
- Memory bandwidth
- 7.70 TB/s
- Architecture
- Blackwell
- Launched
- 2025
These are the chip's own figures, kept in this site's registry and the same whoever rents it out. A provider's feed is trusted for its prices, its GPU counts and the memory it states per card, never for what the board is. There is no speed score on this page: bandwidth is a published number, and turning it into a benchmark would be inventing one.
Every provider renting the B200
Cheapest per GPU-hour first. The rate is for one GPU for one hour, so a node of eight sits beside a single card at a comparable number, with the whole machine's hourly price under it. Where a provider publishes an interruptible rate it is in the same cell.
RunPodB200 · Community Cloud
1x B200 · 180 GB each · 28 vCPU · 283 GB RAM
disk none, billed separately · regions not stated · billed per second · marketplace
Community Cloud: vetted third-party hosts, cheaper and less consistent · storage billed separately · billed per second
$5.98per GPU-hour$5.98/hr all inno spot rate
Verda (DataCrunch)4B200.120V.CC · 4x B200 CC SXM6 180GB
4x B200 · 180 GB each · 120 vCPU · 680 GB RAM · NVLink
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 1.8 TB/s between GPUs
$6.23per GPU-hour$24.93/hr all in$3.12 spot
- Verda (DataCrunch)
8B200.240V.CC · 8x B200 CC SXM6 180GB
8x B200 · 180 GB each · 240 vCPU · 1360 GB RAM · NVLink
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 1.8 TB/s between GPUs
$6.23per GPU-hour$49.86/hr all in$3.12 spot
- Verda (DataCrunch)
2B200.60V · 2x B200 SXM6 180GB
2x B200 · 180 GB each · 60 vCPU · 340 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 900 GB/s between GPUs
$6.42per GPU-hour$12.84/hr all in$3.21 spot
- Verda (DataCrunch)
4B200.120V · 4x B200 SXM6 180GB
4x B200 · 180 GB each · 120 vCPU · 680 GB RAM · NVLink
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 1.8 TB/s between GPUs
$6.42per GPU-hour$25.68/hr all in$3.21 spot
- Verda (DataCrunch)
1B200.30V · 1x B200 SXM6 180GB
1x B200 · 180 GB each · 30 vCPU · 170 GB RAM
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately
$6.42per GPU-hour$6.42/hr all in$3.21 spot
- Verda (DataCrunch)
8B200.240V · 8x B200 SXM6 180GB
8x B200 · 180 GB each · 240 vCPU · 1360 GB RAM · NVLink
disk none, billed separately · regions not stated · billed per hour
Dedicated Hardware Instance · storage billed separately · 1.8 TB/s between GPUs
$6.42per GPU-hour$51.37/hr all in$3.21 spot
Lambda8x NVIDIA B200 SXM6
8x B200 · 180 GB each · 208 vCPU · 2900 GB RAM · NVLink
disk 22 TB · regions not stated · billed per minute
Priced per GPU-hour, billed by the minute · storage included in the instance
$6.69per GPU-hour$53.52/hr all inno spot rate
- RunPod
B200 · Secure Cloud
1x B200 · 180 GB each · 28 vCPU · 283 GB RAM
disk none, billed separately · regions not stated · billed per second
Secure Cloud: RunPod's own datacentres · storage billed separately · billed per second
$6.79per GPU-hour$6.79/hr all inno spot rate
- Lambda
4x NVIDIA B200 SXM6
4x B200 · 180 GB each · 104 vCPU · 1440 GB RAM · NVLink
disk 11 TB · regions not stated · billed per minute
Priced per GPU-hour, billed by the minute · storage included in the instance
$6.79per GPU-hour$27.16/hr all inno spot rate
- Lambda
2x NVIDIA B200 SXM6
2x B200 · 180 GB each · 52 vCPU · 720 GB RAM · NVLink
disk 5.5 TB · regions not stated · billed per minute
Priced per GPU-hour, billed by the minute · storage included in the instance
$6.89per GPU-hour$13.78/hr all inno spot rate
- Lambda
1x NVIDIA B200 SXM6
1x B200 · 180 GB each · 26 vCPU · 360 GB RAM
disk 2.8 TB · regions not stated · billed per minute
Priced per GPU-hour, billed by the minute · storage included in the instance
$6.99per GPU-hour$6.99/hr all inno spot rate
Together AINVIDIA HGX B200
1x B200 · 180 GB each · host size not stated
disk none, billed separately · regions not stated · billed per hour
GPU Clusters, on demand and per GPU-hour · shared filesystem billed separately · reserved terms of 7 days and longer cost less and are not indexed · memory is the board's published spec, which Together's table does not give
$8.19per GPU-hour$8.19/hr all in$4.09 spot
Vultrvbm-256c-3072gb-8-b200-gpu
8x B200 · 192 GB each · 128 vCPU · 3072 GB RAM
disk 3.6 TB · regions not stated · billed per hour
Bare metal · $45696/mo if kept a whole month
$8.50per GPU-hour$68/hr all inno spot rate
CoreWeaveNVIDIA HGX B200
8x B200 · 180 GB each · 128 vCPU · 2048 GB RAM · NVLink
disk 61.4 TB · 2 regions · billed per hour
$68.8 an hour for the whole node · egress and VPC free · IPv4 and storage billed separately · spot shown is the cheaper of the two regions
$8.60per GPU-hour$68.80/hr all in$4.26 spot
Billed as a function
2 rates · never rankedThese rates buy a container that runs when it is called, not a machine you hold. CPU and memory are billed on top, a chosen region or a guarantee against preemption costs more, and the figure is a floor rather than a price. They are listed for what they are and ranked nowhere on this site.
ModalNvidia B200
1x B200 · 180 GB each · host size not stated
disk none, billed separately · regions not stated · billed per second · serverless
$0.001736 a GPU-second, the preemptible region-agnostic rate · a chosen region costs 1.15x to 1.75x and non-preemptible 3x · CPU, memory and volumes billed on top · memory is the board's published spec
$6.25per GPU-hour$6.25/hr all inno spot rate
BasetenB200 180 GiB
1x B200 · 180 GB each · host size not stated
disk not stated · regions not stated · billed per minute · serverless
$0.16633 a GPU-minute on a dedicated deployment · CPU instances priced separately · Baseten publishes no regions, storage or egress rate beside this
$9.98per GPU-hour$9.98/hr all inno spot rate
Frequently asked
- Who rents the B200 cheapest?
- RunPod, at $5.98/GPU-hr on B200 · Community Cloud, read on 17 Sept 2026. That is the on-demand rate for one GPU for one hour; the whole machine bills $5.98/hr.
- What is the NVIDIA B200 SXM?
- The training board of the Blackwell generation, eight to a node with NVLink between them. It is a Blackwell board from 2025 with 180 GB of memory at 7.70 TB/s, in the SXM package. Those figures come from the vendor, not from any provider's price page.
- Can I rent a B200 on spot?
- Yes. 8 of the 15 configurations here publish an interruptible rate, the cheapest being $3.12/GPU-hr at Verda (DataCrunch), 50% under its own on-demand rate. The machine can be taken back mid-run, so it suits work that checkpoints.
- Why do B200 prices differ so much?
- The spread here runs from $5.98/GPU-hr to $8.60/GPU-hr for the same silicon, because what else comes with it differs: the vCPU and RAM around the GPU, whether a disk is included, how many regions it runs in, and whether the machine belongs to the provider or to a third party on a marketplace. Each row says which.
- How current are these prices?
- They are read every morning from each provider's own public page or API, and this page recomputes with them. Last refresh: 17 Sept 2026. Reserved and committed rates are not indexed. Prices exclude tax.