Neutral cross-platform price index: 9 GPU models × 6 platforms (RunPod / ThunderCompute / AtmosCompute + Vast.ai / Lambda / AWS benchmarks). Manually verified 2026-09-23. JSON, CORS-enabled, no vendor lock-in.
| Endpoint | Description | Access |
|---|---|---|
| GET /v1/gpus | Full price table, 9 GPU models × platforms. Query: ?model=A100 ?platform=runpod | Free |
| GET /v1/cheapest | Best price & savings per model | Free |
| GET /v1/platforms | Platform directory | Free |
| GET /v1/futures/index | Live OEX-GCI compute futures index — 23 platforms × 14 GPU models, H100-equivalent normalized, bid/ask & spread | Free |
| GET /v1/futures/components | Futures index full components — per-GPU spot prices & weights | Free |
| GET /v1/exchange/book | Spot matching order book — live bids/asks & recent matches | Free |
| GET /v1/history | Snapshot history & market benchmarks | Pro+ |
| GET /v1/health | Service status | Public |
curl "https://a959e52db0499cef3.app.workbuddy.host/v1/cheapest?model=H100" # live compute futures index (free, no key) curl "https://a959e52db0499cef3.app.workbuddy.host/v1/futures/index" # with API key (Pro) curl -H "X-API-Key: YOUR_KEY" \ "https://a959e52db0499cef3.app.workbuddy.host/v1/history"
{
"model": "RTX 4090",
"best_platform": "atmos",
"best_price_usd_hr": 0.14,
"saving_pct_vs_dearest": 81.1,
"all_on_demand": { "runpod": 0.74, "atmos": 0.14, "vast": 0.29, "lambda": 0.5 }
}