- Interruptible
- $1.38/hr
- Reserved
- Unavailable
The world's cheapestGPU compute.
Prepaid interruptible H100, H200, and B300 GPU-hours in the US and EU.
Interruptible
H100 interruptible
$1.38/ GPU-hour
Updated Oct 31, 2026 · delayed
H100 Compute Cheap interruptible price $1.38 per GPU-hour on Oct 31, 2026MarketCompute Cheap price
Select H100, H200, or B300 interruptible capacity. B200 is sold out. Hover or focus the chart to compare the daily market benchmark with the Compute Cheap price. When the chart is focused, use the left and right arrow keys to inspect dates.
Trusted by teams training and serving frontier models.
Get cheap compute
Midjourney
“We train and merge small language models on interruptible H100s. Preemptions are rare enough that normal checkpointing covers them, and the per-run cost came in well under what we paid before.”
Infra @ Arcee AI
Lowest-price GPU capacity
Limited interruptible capacity.
How it works
From request to running in three steps.
Step 1: Request capacity
Choose H100, H200, or B300 interruptible capacity, and how many GPU-hours you need. Minimum reservation is 2,000 GPU-hours.
Step 2: We confirm availability
Capacity is limited, so we confirm it first. You prepay 50% upfront at the quoted rate, and your interruptible block is booked.
Step 3: Run your workload
Run your workload at the quoted interruptible rate. Capacity can be reclaimed, so save checkpoints.
FAQ
Questions, answered.
Is reserved capacity available?
New compute requests are currently interruptible only. Capacity can be reclaimed if demand spikes, so it suits fault-tolerant training and batch jobs. Reserved capacity is unavailable for new requests; existing agreed terms stay in place.
Is there a minimum reservation?
Yes, 2,000 GPU-hours per request. That's roughly 8 GPUs for 10 days, or fewer GPUs over a longer window, split however your workload needs.
How does the 50% deposit work?
We confirm availability for your exact request before you pay anything. Once confirmed, you prepay 50% of the block upfront to reserve it, and the remaining 50% is billed as usage completes.
What happens if the capacity I asked for isn't available?
We only take a deposit after availability is confirmed, so you're never charged for a block we can't deliver. If exact availability is tight, we'll offer the closest match before you commit.
Which regions is capacity available in?
United States and European Union today. Mention your preferred region in your request's notes, and we'll confirm availability there before anything is reserved.
How is the price calculated?
Hourly rate × GPU-hours, shown before you request. There's no separate platform fee layered on top of the GPU-hour price.
capacity.route()
/ . \ / .\ :. .: .
gpu.alloc() . / \ .queue.next()
.\ . : : . / \
H200 >-------==>-------- [ POOL 01 ] ---==>-----==>--- TRAIN
spot | demand
H100 -==>------==>------ [ POOL 02 ] ==>-----==>------ INFER
spot | demand
: : . RUN > PAY > RELEASE . /\ .
. /\ . / compute.cheap/ \
. \ . . : . : .