RunPod operates a dual-tier marketplace: community GPUs at ultra-low spot prices and a SOC 2-compliant Secure Cloud for production inference. Per-second billing, instant provisioning, and a broad catalog spanning H100, A100, RTX, and even MI300X accelerators make it flexible for projects of any scale.
AMD Instinct MI300X cloud pricing
- Vendor
- AMD
- VRAM
- 192 GB
- Architecture
- CDNA 3
- FP16
- 1307 TFLOPS
- Launched
- 2023
Lowest
$1.19
Median
$1.62
Highest
$2.50
7 results
| Provider | Plan | Price | Regions | Visit | ||||
|---|---|---|---|---|---|---|---|---|
|
|
MI300X (marketplace) On-demand | 1 | 192 GB | 16 | 128 GB |
$1.99
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
MI300X (marketplace) Spot | 1 | 192 GB | 16 | 128 GB |
$1.19
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
MI300X (marketplace) Reserved | 1 | 192 GB | 16 | 128 GB |
$1.29
/GPU-hr
Verified
|
3 countries | Visit → |
|
|
MI300X (Secure Cloud) On-demand | 1 | 192 GB | 24 | 240 GB |
$2.49
/GPU-hr
Verified
|
1 country | Visit → |
|
|
MI300X (Secure Cloud) Reserved | 1 | 192 GB | 24 | 240 GB |
$1.62
/GPU-hr
Verified
|
1 country | Visit → |
|
|
MI300X (per GPU) On-demand | 1 | 192 GB | 24 | 240 GB |
$2.50
/GPU-hr
Verified
|
1 country | Visit → |
|
|
MI300X (per GPU) Reserved | 1 | 192 GB | 24 | 240 GB |
$1.62
/GPU-hr
Verified
|
1 country | Visit → |
Providers offering this GPU
Crusoe is a climate-aligned GPU cloud that runs H100, H200, and MI300X workloads on stranded or flare-captured energy, drastically reducing the carbon footprint of AI compute. For teams that care about sustainability without sacrificing performance, it offers enterprise-grade infrastructure with genuine environmental accountability.
Vast.ai is a decentralized GPU marketplace where host operators list idle accelerators at market-driven prices - often the lowest spot rates anywhere. The trade-off is variable reliability, no compliance certs, and community-grade support; ideal for cost-sensitive experimentation, fine-tuning, and batch inference where occasional interruption is acceptable.
Frequently asked questions
How much does an AMD Instinct MI300X cost per hour in the cloud?
The lowest on-demand AMD Instinct MI300X price we track is $1.19 per GPU-hour. Spot and reserved rates are usually lower; sort the table above by price to see the current rate from every provider.
What is the cheapest AMD Instinct MI300X cloud provider?
Sort the table by price (low to high) to see the cheapest AMD Instinct MI300X provider right now. Marketplace and spot providers often undercut hyperscalers by a wide margin for the same AMD Instinct MI300X.
Which cloud providers offer AMD Instinct MI300X GPUs?
Every provider with published AMD Instinct MI300X availability is listed above, with per-hour pricing, the number of GPUs per instance, region coverage, and on-demand, spot, and reserved rates.
Is spot AMD Instinct MI300X cheaper than on-demand?
Yes. Spot (preemptible) capacity is typically 40-70% cheaper than on-demand but can be reclaimed at short notice. Use the pricing-mode filter to compare on-demand, spot, and reserved rows side by side.
How much VRAM does the AMD Instinct MI300X have?
The AMD Instinct MI300X ships with 192 GB of VRAM. Larger VRAM lets you fit bigger models and batch sizes without sharding.
Is the AMD Instinct MI300X good for AI training and inference?
The AMD Instinct MI300X is used for both LLM training and inference. Match its VRAM and throughput (shown above) to your model size, and use spot capacity for fault-tolerant training to cut costs.