Indices
Each row carries two states. Series tier reflects the live data state per methodology §3.9 — Published when the eligibility window has stabilised, Provisional while a tier-promotion window is still maturing, Raw-only for series tracked but not yet admissible. Methodology links to the methodology document itself — Locked v1.0 means the document is final and citations against this index are stable; Draft v1.0 means the document is still being drafted and values are not yet citable. Only rows whose methodology is Locked v1.0 AND whose series tier is Published can be cited as a Pulse headline figure.
| Index | Series tier | Latest value | Methodology | Description |
|---|---|---|---|---|
| Pulse Inference Token Index | Provisional | — | Locked v1.0 | Median price per million tokens for the commodity-host basket of open inference endpoints. |
| Pulse H100 SXM Hyperscaler OD | Published | — | Locked v1.0 | Median on-demand list price per GPU-hour for H100 SXM across hyperscaler providers. |
| Pulse H100 SXM Neocloud OD | Published | — | Locked v1.0 | Median on-demand list price per GPU-hour for H100 SXM across neocloud providers. |
| Pulse A100 80GB Hyperscaler OD | Published | — | Locked v1.0 | Median on-demand list price per GPU-hour for A100 80GB across hyperscaler providers. |
| Pulse A100 80GB Neocloud OD | Published | — | Locked v1.0 | Median on-demand list price per GPU-hour for A100 80GB across neocloud providers. |
| Pulse B200 Neocloud OD | Draft v1.0 | — | Draft v1.0 | Median on-demand list price per GPU-hour for B200 across neocloud providers. |
| Pulse H200 141GB Neocloud OD | Draft v1.0 | — | Draft v1.0 | Median on-demand list price per GPU-hour for H200 141GB across neocloud providers. |
| Pulse H100 PCIe Neocloud OD | Draft v1.0 | — | Draft v1.0 | Median on-demand list price per GPU-hour for H100 PCIe across neocloud providers. |
| Pulse A100 80GB Neocloud Spot | Draft v1.0 | — | Draft v1.0 | Median spot price per GPU-hour for A100 80GB across neocloud providers. |