Training runs
A job that needs cards for hours and nothing for the rest of the week.
- GPU sizes named from the list the API returns
- GPU-hours metered while the job runs
- Cancel and the hours used so far are billed, then the job stops
Hanzo Machines
High-Performance Computing
Rent a machine by the hour, metered to your organization.
{ "gpu": "1x H100", "dryRun": true }$3.48/ h
80 GB
- an hour
- $3.48
- a working day
- $27.84
- all month
- $2,540
On demand, at the price list’s rate as it stands.
| 1x H100 | 80 GB | $3.48 / h | $27.84 | $2,540 |
| 2x H100 | 160 GB | $6.96 / h | $55.68 | $5,081 |
| 4x H100 | 320 GB | $13.92 / h | $111 | $10,162 |
NVIDIA H100
Dedicated GPU instances for AI/ML training
- VRAM
- 80GB HBM3
- Cores
- 16,896 CUDA
- Tensor Cores
- 528 TCs
AMD EPYC 7763
High-core count CPU instances
- Cores
- 64 Cores
- Threads
- 128 Threads
- Memory
- Up to 2TB
Storage Options
Ultra-fast NVMe and persistent storage
- Local NVMe
- Up to 8TB
- Network Storage
- Unlimited
- I/O Performance
- 7GB/s
When you want a whole machine
Reach for one when the work outlasts a request, or needs a card of its own
Serving a model
A machine that stays up because something is calling it.
- Run Hanzo Engine on a card you rented
- Terminate it when the traffic goes away
- The bill follows the machine, not a reservation
Work that will not fit in a request
Renders, batches and simulations that take longer than an HTTP call.
- A queue per GPU, plus a shared any-GPU lane
- Each item says which node claimed it
- Read the queue instead of guessing where a job went
Agents that need a computer
An agent with a machine under it, launched and torn down as one thing.
- Bind a cloud agent to a machine
- Message it, or stop it, by name
- Tearing down the pair unbinds the agent first
How renting one works
Every step is an API call, so anything you do once by hand you can put in a script afterwards
- 01
Ask what exists
The sizes and regions you can launch into are an endpoint, not a page in the docs. Read them, then pick one — nothing to keep in sync by hand.
- 02
Price it before you buy it
A launch with dryRun returns the price and creates nothing. The same call without it creates the machine, so what you were quoted is what you asked for.
- 03
Metered to the organization
Time is debited from your org's prepaid balance while a machine is up. Before it launches, the cloud says whether you may — and what is missing if you may not.
- 04
One list, whoever owns them
Rented machines, the droplets and cluster workers behind them, and your own hardware that dialed in with hanzo link all appear as one fleet, deduplicated.
- 05
A row per accelerator
The GPU view counts real cards: from the size of the nodes you rented, and from what your own workers report through nvidia-smi. Where there is no inventory, it returns nothing rather than inventing something.
- 06
A machine can carry an agent
Bind a cloud agent to a machine and the pair is one thing to launch, message, stop and tear down. Unbinding is separate, so the agent can move without the machine going with it.
Whose machine it is
Ownership, region and account are decisions the API makes you state
01
Scoped to your organization
A machine belongs to one org and is named within it. Every read and every terminate is checked against the org on your token, not against the id in the URL.
02
Bring your own account
Link a DigitalOcean, AWS or GCP account and the clusters it holds fold into the same fleet. Your capacity, your contract, one place to see it.
03
Pick the region deliberately
Launches name a region from the list the API returns, so where a workload runs is a decision you made rather than one that was made for you.
Paste and ship
Paste this into any agent. It reads the skill manifest and calls Machines from there.
api /v1/compute/machines
Read https://hanzo.ai/skill.md and use Hanzo Machines in my project. Start with: hanzo visor machines list