**Persistent AI demand for large language model training and inference continues to support H100 rental rates near $2.50–$3.00 per GPU-hour as September closes.** Neocloud providers such as RunPod, Vast.ai, and Lium offer on-demand capacity from roughly $1.30–$2.90, while hyperscalers like AWS and Azure command $6–$12 premiums for integrated ecosystems and reliability. Trackers show the median on-demand price around $3.20–$3.35, up modestly over the past 90 days amid tightening supply and Blackwell rollout delays that sustain Hopper utilization. Spot and interruptible options trade lower, creating a wide but stable distribution. With only days left, last-minute capacity releases or inference spikes represent the main variables that could shift the final observed rate within the closely matched leading buckets.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated$2.50-$2.75 25%
$3.00-$3.25 23.8%
$2.75-$3.00 17%
$2.25-$2.50 16.0%
$41,412 Vol.
$41,412 Vol.
<$2.25
5%
$2.25-$2.50
16%
$2.50-$2.75
25%
$2.75-$3.00
17%
$3.00-$3.25
14%
$3.25+
8%
$2.50-$2.75 25%
$3.00-$3.25 23.8%
$2.75-$3.00 17%
$2.25-$2.50 16.0%
$41,412 Vol.
$41,412 Vol.
<$2.25
5%
$2.25-$2.50
16%
$2.50-$2.75
25%
$2.75-$3.00
17%
$3.00-$3.25
14%
$3.25+
8%
If the recorded data falls exactly between two brackets, this market will resolve to the higher bracket.
The resolution source for this market is Ornnai.com (ornnai.com), specifically, the H100 Index chart data available at https://dashboard.ornnai.com. The specified finalized daily value shown on the chart will be used for resolution. Daily data will be considered finalized once the following day’s data point is published.
Resolution will occur once the specified data point is finalized. If the relevant data is not finalized by the end of the 7th calendar day after the specified date (ET), this market will resolve according to the latest data available at that time. Revisions made after the relevant figure has been finalized will not be considered.
Market Opened: Jul 30, 2026, 11:28 AM ET
Resolution Source
https://dashboard.ornnai.com/computeResolver
0x69c47De9D...If the recorded data falls exactly between two brackets, this market will resolve to the higher bracket.
The resolution source for this market is Ornnai.com (ornnai.com), specifically, the H100 Index chart data available at https://dashboard.ornnai.com. The specified finalized daily value shown on the chart will be used for resolution. Daily data will be considered finalized once the following day’s data point is published.
Resolution will occur once the specified data point is finalized. If the relevant data is not finalized by the end of the 7th calendar day after the specified date (ET), this market will resolve according to the latest data available at that time. Revisions made after the relevant figure has been finalized will not be considered.
Resolution Source
https://dashboard.ornnai.com/computeResolver
0x69c47De9D...**Persistent AI demand for large language model training and inference continues to support H100 rental rates near $2.50–$3.00 per GPU-hour as September closes.** Neocloud providers such as RunPod, Vast.ai, and Lium offer on-demand capacity from roughly $1.30–$2.90, while hyperscalers like AWS and Azure command $6–$12 premiums for integrated ecosystems and reliability. Trackers show the median on-demand price around $3.20–$3.35, up modestly over the past 90 days amid tightening supply and Blackwell rollout delays that sustain Hopper utilization. Spot and interruptible options trade lower, creating a wide but stable distribution. With only days left, last-minute capacity releases or inference spikes represent the main variables that could shift the final observed rate within the closely matched leading buckets.
Experimental AI-generated summary referencing Polymarket data. This is not trading advice and plays no role in how this market resolves. · Updated
Beware of external links.
Beware of external links.
Frequently Asked Questions