GPU comparison
MI355X vs H200: Specs, monthly price and which to rent
MI355X is a generation ahead of the H200: about twice the memory (288 vs 141 GB), 1.7× the bandwidth (8 vs 4.8 TB/s), 2.5× the dense FP8 (5.0 vs 1.98 PF) and FP4 that Hopper does not have, for about 9% less per month. The H200 stays the safe choice when your code depends on CUDA-only kernels, TensorRT-LLM or recipes tuned for Hopper.
- Updated Nov 4, 2026
- Specs: vendor datasheets, dense Tensor figures
- Prices: live CoreGPU catalog
| Spec | AMD Instinct MI355X | NVIDIA H200 SXM |
|---|---|---|
| Architecture | CDNA 4 | Hopper |
| GPU memory | 288 GB HBM3E | 141 GB HBM3e |
| Memory bandwidth | 8 TB/s | 4.8 TB/s |
| FP8 dense | 5 PF | 1.98 PF |
| FP4 dense | 10.1 PF | n/a |
| Interconnect | Infinity Fabric · 1.08 TB/s | NVLink 4 · 900 GB/s |
| Power (TDP) | 1,400 W | 700 W |
| Price per GPU / month | $2,449 | $2,699 |
| Price per GB of memory | $8.50 | $19.14 |
| Price per FP8 PFLOPS | $490 | $1,363 |
| Availability | 21 available | 34 available |
Green marks the better value on each line. Prices are per GPU per month at CoreGPU, at least 25% below the market median.
- Serving 70B to 400B models on fewer GPUs
- FP8 and FP4 inference on ROCm
- More memory and compute per dollar
- CUDA-only kernels and TensorRT-LLM
- Recipes tuned for Hopper
- NVIDIA tooling across the whole team
MI355X vs H200: monthly price by server size
| Server | MI355X | H200 | Difference |
|---|---|---|---|
| 1 GPU | $2,449/mo | $2,699/mo | $250 less on the MI355X |
| 2 GPUs | $4,898/mo | $5,398/mo | $500 less on the MI355X |
| 4 GPUs | $9,796/mo | $10,796/mo | $1,000 less on the MI355X |
| 8 GPUs | $19,592/mo | $21,592/mo | $2,000 less on the MI355X |
Flat monthly price per GPU, the same at every server size. USD, taxes excluded.
Largest model that fits in GPU memory
| Weights | 1× MI355X | 1× H200 | 8× MI355X | 8× H200 |
|---|---|---|---|---|
| BF16 | 120B | 58B | 960B | 470B |
| FP8 | 240B | 117B | 1.9T | 940B |
| FP4 | 480B | 235B | 3.8T | 1.9T |
Parameters whose weights fit with 20% headroom for activations and runtime. Long contexts and large batches need more room for the KV cache: see how much GPU memory you need.
More GPU comparisons
MI355X vs H200: common questions
Already a customer? Open a ticket from the console: engineers reply within one business day.
How support worksIs the MI355X faster than the H200?
On dense FP8 Tensor throughput the MI355X is faster: 5 PF against 1.98 PF per GPU. Memory bandwidth is 8 TB/s on the MI355X and 4.8 TB/s on the H200, which matters most for LLM decoding.
How much does the MI355X cost per month compared with the H200?
At CoreGPU the MI355X is $2,449 and the H200 is $2,699 per GPU per month, so an 8-GPU server costs $19,592 or $21,592 a month. The MI355X is 9% cheaper per GPU. Both are at least 25% below the market median, billed monthly from a prepaid balance.
Which is better for LLM inference, the MI355X or the H200?
For large models and long contexts the MI355X has the edge with 288 GB of memory per GPU, so a model needs fewer GPUs. Choose the MI355X for serving 70B to 400B models on fewer GPUs; the H200 for cUDA-only kernels and TensorRT-LLM.
Can I rent the MI355X and the H200 by the month?
Yes. Both are dedicated GPUs you rent by the month, from 1 to 8 per server, with root SSH access. Right now the MI355X is 21 available and the H200 is 34 available. You pay in BTC, ETH, XMR, USDT (TRC-20) or SOL.