Andromeda Cluster: 15 exaflops^* for startups

• 3,200 H100s on 400 nodes interlinked with 3.2Tbps infiniband
• 432 H100s on 54 nodes interlinked with 3.2Tbps infiniband
• 768 A100s for training and inference with 1.6Tbps infiniband
• Available for experiments, training runs, and inference
• You can queue training runs that use the entire cluster, or part of it, or just ssh in
• Store data locally on NAS, or stream it in; no ingress/egress costs
• No minimum duration and superb pricing
• Big enough to train llama 65B in a week
• Total mass of 4,862 kg (GPUs only; not counting chassis, system, rack)
• For use by startup investments of Nat Friedman and Daniel Gross
• Reach out if you don't have access, or login if you already have it

We also operate a marketplace for third-party GPUs at gpulist.ai.

^* based on 3,958 teraFLOPS per H100 at FP8 with sparsity, your results may vary, consult your doctor before beginning a new training routine