- Home
- Datacenters
- Inflection-2 training cluster
Existing Likely
Inflection-2 training cluster
Inflection AI,CoreWeave • United States of America • July 30, 2023
Total GPUs
5.0K
H100 Equiv
5.0K
Power
7.3 MW
FP16
4.9K PF
System Equivalents
This datacenter's 5.0K GPUs equal:
NVIDIA DGX H100
8 GPUs/system
625
NVIDIA HGX H100 (8-GPU)
8 GPUs/system
625
NVIDIA HGX H100 (4-GPU)
4 GPUs/system
1.3K
Specifications
| Primary GPU | NVIDIA H100 SXM5 80GB |
| GPU Count | 5.0K |
| Sector | Private |
Additional Information
Inflection-2 was trained on 5,000 NVIDIA H100 GPUs
Inflection lent 3,500 H100 GPUs to Corweave (who appears to be the party that built that cluster) for the June MLPerf submission. Given that this is just slightly larger and operational just after, it appears very likely that this cluster is an extension of the MLPerf cluster (or the MLPerf cluster was just a subset of this cluster). It's possible that both of these subsets of the Eos-DFW cluster, but there isn't much evidence for this besides that Eos-DFW was submitted to MLPerf by CoreWeave later that year (I give it a ~50% chance this is Eos-DFW)