Open protocol and registry for discovering, verifying and pricing AI inference e
Open protocol and registry for discovering, verifying and pricing AI inference endpoints. MCP server, OpenAI-compatible gateway, Ollama router, and a Node Kit for operators.
active Production-safe Score 96/100 http no auth
At a glance
| Endpoint | https://registry.opennodes.io/mcp |
|---|---|
| Source | https://github.com/opennodes-io/opennodes |
| Transport | http |
| Tools | 7 |
| Auth | none |
| First seen | 2026-11-02 21:53:34 |
| Last checked | 2026-11-17 21:50:55 |
Observed reliability
| 7-day uptime | 100% |
|---|---|
| 24-hour uptime | 100% |
| Handshake success | 100% |
| Latency p50 / p95 | 245 / 426 ms |
| Last successful check | 2026-11-17 21:50:55 |
Measured by Arclan's validator via repeated live MCP handshakes. How this is computed →
Tools (7)
recommend— Recommend the best AI node + model for a task, with an enforceable cost estimate and human-readable reasons. Use this first when you know thsearch_offerings— Search AI inference offerings in the OpenNodes registry — registered nodes AND imported catalogs (OpenRouter, Hugging Face Inference Provideget_offering— Full detail for one offering: endpoints, pricing, serving capabilities (context window, supported features, languages, hardware), data policget_node— Who runs a node and how it has behaved: operator, country and regions, endpoints, declared hardware, trust tier and its history (verificatioestimate— Pre-price an invocation chain BEFORE spending money. Each step is either a concrete offering ("node_id/offering_id") or a `select` query thafind_locations— Where AI models are served, by country: every known serving site (provider, region, city or facility, live or planned) with its evidence basfind_residency— What each AI provider officially promises about where prompts, outputs and stored data stay: per provider (cloud, lab API, inference host or
Validation history
| Checked | Reachable | Handshake | Latency | Score | State |
|---|---|---|---|---|---|
| 2026-11-17 21:50:55 | ✓ | ✓ | 134 ms | 95 | active |
| 2026-11-17 21:35:43 | ✓ | ✓ | 118 ms | 95 | active |
| 2026-11-17 21:20:26 | ✓ | ✓ | 118 ms | 95 | active |
| 2026-11-17 21:05:17 | ✓ | ✓ | 153 ms | 95 | active |
| 2026-11-17 20:49:55 | ✓ | ✓ | 291 ms | 95 | active |
| 2026-11-17 20:34:45 | ✓ | ✓ | 272 ms | 95 | active |
| 2026-11-17 20:19:38 | ✓ | ✓ | 136 ms | 95 | active |
| 2026-11-17 20:04:19 | ✓ | ✓ | 116 ms | 95 | active |
| 2026-11-17 19:49:06 | ✓ | ✓ | 283 ms | 95 | active |
| 2026-11-17 19:34:00 | ✓ | ✓ | 279 ms | 95 | active |
| 2026-11-17 19:18:46 | ✓ | ✓ | 167 ms | 95 | active |
| 2026-11-17 19:03:31 | ✓ | ✓ | 117 ms | 95 | active |
| 2026-11-17 18:48:17 | ✓ | ✓ | 268 ms | 95 | active |
| 2026-11-17 18:33:03 | ✓ | ✓ | 117 ms | 95 | active |
| 2026-11-17 18:17:51 | ✓ | ✓ | 121 ms | 95 | active |
Query this programmatically
Agents can pull this record live over the MCP protocol or REST:
get_server("https://registry.opennodes.io/mcp") # via the Arclan MCP endpoint https://arclan.ai/mcp
GET https://arclan.ai/api/registry/registry-opennodes-io-mcp # JSON