NVIDIA B200 cloud prices
B200 on-demand cloud prices run from $8.50 to $14.24 per GPU-hour across 4 providers gpu.gl tracks (median $11.57). The lowest is Nebius.
Lowest price by provider
| Configuration | Where | Fetched | Source | |||
|---|---|---|---|---|---|---|
| Nebius | On-demand | $8.50 | HGX B200 · 1 GPU | Not stated per region | Nebius ↗ | |
| DataCrunch | On-demand | $9.14 | 4B200.120V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ | |
| OCI | On-demand | $14.00 | B110978 (Compute - GPU - B200) · 1 GPU | All commercial regions | OCI ↗ | |
| AWS | On-demand | $14.24 | p6-b200.48xlarge · 8 GPUs | 2 regions | AWS ↗ | |
| DataCrunch | Spot / interruptible | $4.57 | 8B200.240V · 1, 2, 4, 8 GPUs | Not stated per region | DataCrunch ↗ |
Price history
History starts 2026-11-08. gpu.gl stores one snapshot a day from launch and never back-fills, so a chart appears here once there are two daily snapshots with on-demand prices for this GPU (1 so far).
B200 specifications
The B200 has 180 GB of HBM3E and 7.7 TB/s of memory bandwidth. That is about 292 dense FP16 TFLOPS per TB/s of bandwidth; workloads that stream large weights, like LLM inference at small batch sizes, tend to be limited by the bandwidth figure, while large-batch training leans on the compute figure.
| Architecture | Blackwell |
|---|---|
| Memory | 180 GB HBM3E |
| Memory bandwidth | 7.7 TB/s |
| FP16 / BF16 tensor (dense) | 2,250 TFLOPS |
| FP8 tensor (dense) | 4,500 TFLOPS |
| Max power | 1000 W |
| Form factor | SXM |
Source: NVIDIA spec sheet. Tensor figures are dense (without sparsity). Per-GPU figures from the HGX B200 datasheet. NVIDIA’s DGX B200 page implies 8 TB/s per GPU.
Every configuration
Each instance size or listing, at its cheapest region.
| Provider | Price type | $/GPU-hr | Instance | GPUs | Instance price | Cheapest region | Fetched | Source |
|---|---|---|---|---|---|---|---|---|
| Nebius | On-demand | $8.50 | HGX B200 | 1 | $8.50 per GPU-hour | Not stated per region | Nebius ↗ | |
| DataCrunch | On-demand | $9.14 | 4B200.120V | 4 | $36.55/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | On-demand | $9.14 | 8B200.240V | 8 | $73.10/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | On-demand | $9.14 | 1B200.30V | 1 | $9.14/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | On-demand | $9.14 | 2B200.60V | 2 | $18.28/hr | Not stated per region | DataCrunch ↗ | |
| OCI | On-demand | $14.00 | B110978 (Compute - GPU - B200) | 1 | $14.00 per GPU-hour | All commercial regions | OCI ↗ | |
| AWS | On-demand | $14.24 | p6-b200.48xlarge | 8 | $113.93/hr | US East (N. Virginia) +1 more | AWS ↗ | |
| DataCrunch | Spot / interruptible | $4.57 | 8B200.240V | 8 | $36.55/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | Spot / interruptible | $4.57 | 1B200.30V | 1 | $4.57/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | Spot / interruptible | $4.57 | 2B200.60V | 2 | $9.14/hr | Not stated per region | DataCrunch ↗ | |
| DataCrunch | Spot / interruptible | $4.57 | 4B200.120V | 4 | $18.28/hr | Not stated per region | DataCrunch ↗ |