Two Nvidia data center GPUs, one Blackwell Ultra and one Hopper: 288 GB HBM3e against 141 GB HBM3e and 67% more memory bandwidth on the B300.
At a glance
Memory
- B300
- 288 GB
- H200
- 141 GB
2x more on B300
Memory bandwidth
- B300
- 8.00 TB/s
- H200
- 4.80 TB/s
1.7x higher on B300
Cheapest on-demand
- B300
- $6.50 /GPU/hr
- H200
- $2.09 /GPU/hr
3.1x more expensive on B300
Price comparison
Median price per GPU per hour by billing type, with the cheapest listing under it.
| Billing | B300 | H200 | Difference |
|---|---|---|---|
| On-demand | $7.85 /GPU/hr from $6.50 | $4.39 /GPU/hr from $2.09 | 44% cheaper on H200 |
| Reserved | $5.71 /GPU/hr from $4.25 | $3.54 /GPU/hr from $2.28 (1 mo) | 38% cheaper on H200 |
| Spot | $4.48 /GPU/hr from $4.30 | $2.62 /GPU/hr from $1.32 | 42% cheaper on H200 |
Heads up: A provider's own page may quote a different figure, for example on tax, region or a promotion. A monthly-only plan shows a derived hourly rate. Medians are across providers, one vote each, so a large catalog does not outweigh a small one. Verify before provisioning. More on how we price.
Price history
B300 vs H200 price history
Weekly median on-demand price per GPU per hour
Specs comparison
| Spec | B300 | H200 | Difference |
|---|---|---|---|
| Memory | |||
| Capacity | 288 GB HBM3e | 141 GB HBM3e | 2x more on B300 |
| Bandwidth | 8,000 GB/s | 4,800 GB/s | 1.7x higher on B300 |
| Bus width | 8,192-bit | ||
| Compute · dense | |||
| FP4 | 14,000 TFLOPS | Not supported | B300 only |
| FP8 | 4,500 TFLOPS | 1,979 TFLOPS | 2.3x higher on B300 |
| FP16 / BF16 | 2,250 TFLOPS | 989.5 TFLOPS | 2.3x higher on B300 |
| INT8 | 153.5 TOPS | 1,979 TOPS | 13x higher on H200 |
| FP32 | 75 TFLOPS | 67 TFLOPS | 1.1x higher on B300 |
| FP64 | 1.2 TFLOPS | 34 TFLOPS | 28x higher on H200 |
| Platform | |||
| Interconnect | NVLink 1,800 GB/s | NVLink 900 GB/s | 2x higher on B300 |
| Board power | 1,100 W | 700 W | 1.6x higher on B300 |
| Released | Q1 2025 | Q4 2023 | 16 months newer on B300 |
| Precision | |||
| Supported formats | FP4FP6FP8FP16BF16TF32FP32FP64INT8 | FP8FP16BF16TF32FP32FP64INT8 | |
Compute: Vendor peak figures, dense and per GPU. Not measured throughput.
Source: B300 datasheet and H200 datasheet.
Where to rent them
Providers listing these GPUs, those with both first, so you can switch without moving clouds.
| Provider | B300 /GPU/hr | H200 /GPU/hr |
|---|---|---|
GPU.ai
|
On-Demand $7.89 | On-Demand $3.29 |
Runpod
|
On-Demand $7.89 | On-Demand $4.59 |
TheAI Cloud
|
Reserved (24mo) $4.90 | -- |
GPUaaS
|
Custom on request | -- |
Packet·ai
|
Custom on request | -- |
Impossible Cloud
|
Reserved (60mo) $4.32 | -- |
Canopy Wave
|
Custom on request | -- |
Bentaus
|
On-Demand $6.50 | -- |
Koyeb
|
-- | On-Demand $3.00 |
Lium
|
-- | On-Demand $3.00 |
Vast.ai
|
-- | On-Demand $3.30 |
Massed Compute
|
-- | On-Demand $3.62 |
Lyceum
|
-- | On-Demand $4.29 |
EmpirioLabs AI
|
-- | On-Demand $5.50 |
30 providers list the B300 and 46 the H200. Every listing is on the B300 page and the H200 page.
Similar GPUs
Other comparisons in the same class.
-
Nvidia H100
From $1.30 /GPU/hr Compare -
Nvidia B200
From $3.75 /GPU/hr Compare -
Nvidia GB300
From $18.00 /GPU/hr Compare
Our data for Nvidia B300 was last updated on Sept. 14, 2026, and for Nvidia H200 on Sept. 14, 2026.