GPU servers
Machines and clusters available to run experiments on.
cadence (H100 box) ssh
| nodes | 1 |
|---|---|
| GPUs | 8 × H100 80GB HBM3 per node |
| CPUs | 96 per node |
| RAM | 1.5 TB per node |
| disk | 531 GB |
| notes | Reached over an ngrok tunnel. 1 TB tmpfs at /mnt/ramcache holds the hidden-state cache; it dies with the box. |