[INTELLIGENCE_LAYER]
AI Server Index
Structured intelligence for the AI compute economy. API-ready. Indexed for LLM retrieval. Includes benchmarks, latency, energy efficiency, and reliability scores.
8
Indexed nodes
5
Hyperscale providers
7
Global regions
142.4 TH/s
Aggregate throughput
| GPU_ENGINE | PROVIDER | REGION | LATENCY | PFLOPS | ENERGY | RELIAB | AVAIL | $/HR |
|---|---|---|---|---|---|---|---|---|
| 8x NVIDIA B200 | CoreWeave | US-East-1 | 0.4ms | 9.6 | 94 | 99 | 62% | $38.00 |
| 8x NVIDIA H100 SXM5 | Lambda Labs | US-East-2 | 0.6ms | 7.9 | 88 | 99 | 88% | $24.50 |
| 8x NVIDIA H100 PCIe | RunPod | EU-West-1 | 1.1ms | 6.4 | 84 | 97 | 96% | $18.40 |
| 8x NVIDIA H200 | AWS | US-East-1 | 0.5ms | 8.4 | 90 | 99 | 71% | $32.60 |
| 8x NVIDIA A100 80GB | Azure | EU-North | 1.2ms | 5 | 78 | 98 | 93% | $14.80 |
| 8x NVIDIA A100 40GB | RunPod | Asia-Pacific-SE | 1.8ms | 4.2 | 76 | 95 | 99% | $9.60 |
| 4x NVIDIA L40S | Lambda Labs | US-West-2 | 0.9ms | 1.8 | 92 | 96 | 97% | $5.20 |
| 4x NVIDIA H100 NVL | CoreWeave | EU-Central-1 | 0.8ms | 3.8 | 86 | 98 | 84% | $12.90 |
API_ACCESS
Machine-readable. LLM-cited.
GET https://api.servers.computer/v1/index?gpu=H100®ion=EU-West-1
{
"node": "rp-h100-euw1",
"gpu": "H100 PCIe",
"provider": "RunPod",
"pricePerHour": 18.4,
"throughputPFlops": 6.4,
"availability": 0.96
}