AMD Instinct MI300X rental pricing
CDNA 3 architecture · 192 GB memory · 5.3 TB/s bandwidth · 750 W board power · Released 2023
Price history
Weekly median price per GPU per hour
At a glance
prices checked 5h ago- Cheapest in stock
- $0.50/hr
- Community · 1× Instinct MI300X 192GB · RunPod
- Median price
- $3.22/hr
- On-demand · 6 providers
- 90-day trend
- Down 6.7%
- Weekly median on-demand, 4 weeks
- Coverage
- 7 providers
- 16 price points
- $ per FP8 PFLOP-hour
- $0.91
- Rank 6 of 31 priced GPUs · lower is better
Every provider we track
Cheapest listing per provider and pricing type. Sold-out rows stay so you can see who normally carries it.
| Provider | Configuration | Price/GPU-hr | Checked | |
|---|---|---|---|---|
| 1× Instinct MI300X 192GB Community | $0.50 | 5h ago | View → | |
| 1× Instinct MI300X 192GB | $2.39 | 5h ago | View → | |
| 1× Instinct MI300X 192GB | $2.59 | 5h ago | View → | |
| 1× Instinct MI300X 192GB | $2.99 | 5h ago | View → | |
 Cirrascale | 8× Instinct MI300X 192GB Reserved | $3.08 | 5h ago | View → |
| 8× Instinct MI300X 192GB Reserved | $3.39 | 5h ago | View → | |
| 1× Instinct MI300X 192GB | $3.45 | 5h ago | View → | |
| 1× Instinct MI300X 192GB | $6.00 | 5h ago | View → | |
| 8× Instinct MI300X 192GB | $6.00 | 5h ago | View → |
Per GPU-hour. Spot, reserved and community pricing differ from on-demand; click through for the provider's current figure.
What can the AMD Instinct MI300X run?
Memory for the weights plus one sequence of context, how many GPUs hold it, and what that replica costs at the cheapest in-stock on-demand rate.
| Model | Params | Memory | GPUs | $/hr |
|---|---|---|---|---|
| gpt-oss-20b OpenAI | 20.9B· 3.6B active | 20 GiB | 1 | $2.39 |
| Gemma 3 27B Google | 27.4B | 26 GiB | 1 | $2.39 |
| Qwen3-32B Alibaba | 32.8B | 32 GiB | 1 | $2.39 |
| Llama 3.3 70B Instruct Meta | 70.6B | 67 GiB | 1 | $2.39 |
| gpt-oss-120b OpenAI | 117B· 5.1B active | 109 GiB | 1 | $2.39 |
| Qwen3-235B-A22B Alibaba | 235B· 22.0B active | 220 GiB | 2 | $4.78 |
| DeepSeek-R1 DeepSeek | 684B· 37.0B active | 638 GiB | 4 | $9.56 |
| Kimi K3 Moonshot AI | 2.78T· 104B active | 2.53 TiB | 16 | $38.24 |
Sized for one sequence at 4K context with 16-bit KV cache and 10% of memory held back. Serving many users at once needs more; the LLM GPU calculator sizes for concurrency and throughput.
AMD Instinct MI300X pricing questions
How much does it cost to rent a AMD Instinct MI300X per hour?
The cheapest AMD Instinct MI300X in stock right now is $0.50 per GPU-hour community from RunPod for the 1× Instinct MI300X 192GB configuration. The median on-demand rate across 6 providers is $3.22 per GPU-hour.
How much does a AMD Instinct MI300X cost per month?
Running one AMD Instinct MI300X around the clock for a month (730 hours) costs about $365 at the cheapest in-stock rate of $0.50 per hour. At the median on-demand rate it is about $2,351 a month. Reserved terms usually come in below that; ask the provider for a quote.
Which provider has the cheapest AMD Instinct MI300X?
RunPod has the cheapest AMD Instinct MI300X in stock right now at $0.50 per GPU-hour community for the 1× Instinct MI300X 192GB configuration. That is a community rate; see the provider table for the cheapest on-demand listing. Rates change often, so check the provider table for today's figure before you commit.
Are AMD Instinct MI300X rental prices going up or down?
The weekly median on-demand rate for the AMD Instinct MI300X is down 6.7% over the last 4 weeks, from $3.45 to $3.22 per GPU-hour.
What is the difference between on-demand, spot and reserved AMD Instinct MI300X pricing?
On-demand is pay-as-you-go with no commitment. Reserved is a committed term, usually months, in exchange for a lower rate. Community is capacity from individual hosts on a marketplace, cheaper and less uniform. This week's median rates for the AMD Instinct MI300X are on-demand $3.22 (6 providers), reserved $3.24 (2 providers), community $0.50 (1 provider), per GPU-hour.
How many cloud providers rent the AMD Instinct MI300X?
We track 7 providers with 16 live prices for the AMD Instinct MI300X. Sold-out listings stay in the table so you can see who normally carries it, but never set the cheapest figure.
What LLMs can you run on one AMD Instinct MI300X?
With 8-bit weights and a single sequence of context, one AMD Instinct MI300X holds gpt-oss-20b (20.9B), Gemma 3 27B (27.4B), Qwen3-32B (32.8B), Llama 3.3 70B Instruct (70.6B), gpt-oss-120b (117B). Kimi K3 at 2.78T needs 16 GPUs at 8-bit. See the model fit table for the full ladder.
How does the AMD Instinct MI300X compare on price per PFLOP-hour?
At its cheapest in-stock on-demand rate the AMD Instinct MI300X delivers dense FP8 compute at $0.91 per PFLOP-hour, which ranks 6 of 31 priced GPUs we cover on that basis. Lower is better: it is the rental price divided by peak FP8 throughput.
Where does this AMD Instinct MI300X pricing data come from?
Prices are read daily from each provider's public price list or API and recorded whenever they change. Listings are shown per GPU-hour on-demand unless marked otherwise, and the chart covers the last 30 days. The newest listing was checked 5h ago.