OpenNodes is an MCP server with 7 tools an agent can call, among them estimate, find_locations, find_residency. It answers at https://registry.opennodes.io/mcp over streamable HTTP, and it listed its tools to an anonymous client, so no sign-in was needed for that. Our probes reached it 2 of 4 times since 5 October 2026; the last check, on 21 November 2026, succeeded.
Discover verified AI inference nodes, get the best node+model per task, pre-price chains.
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
estimate | Pre-price an invocation chain BEFORE spending money. Each step is either a concrete offering ("node_id/offering_id") or a `select` query that picks the best match, plus expected input and output tokens. Returns per-step cost bounds in USD, Inputs: steps |
find_locations | Where AI models are served, by country: every known serving site (provider, region, city or facility, live or planned) with its evidence basis (claimed, documented, measured, attested), its source page, a map point (`geo`: lat, lon, precisi Inputs: country, basis, status, model, exclude_lineage, lineage |
find_residency | What each AI provider officially promises about where prompts, outputs and stored data stay: per provider (cloud, lab API, inference host or router), its commitments, each with a commitment level (processing: inference and stored data stay Inputs: scope, commitment, availability, kind, node_id |
get_node | Who runs a node and how it has behaved: operator, country and regions, endpoints, declared hardware, trust tier and its history (verifications, suspensions, reinstatements, with reasons), recent registry probe results with an availability s Inputs: node_id, offering_limit |
get_offering | Full detail for one offering: endpoints, pricing, serving capabilities (context window, supported features, languages, hardware), data policy, trust tier, and the registry's observed probe results. Take node_id and offering_id from search_o Inputs: node_id, offering_id |
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 the task but not the model. Pass `task` (sent to this registry, which extracts features and discards t Inputs: task, features, est_output_tokens, policy |
search_offerings | Search AI inference offerings in the OpenNodes registry — registered nodes AND imported catalogs (OpenRouter, Hugging Face Inference Providers, models.dev, local nodes). All filters are optional and combine with AND. Returns { offerings, to Inputs: modality, family, min_context, supports, max_input_price, scheme |
Every block below points at https://registry.opennodes.io/mcp. Pick your client and copy.
claude mcp add --transport http opennodes https://registry.opennodes.io/mcp
{
"mcpServers": {
"opennodes": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://registry.opennodes.io/mcp"]
}
}
}{
"mcpServers": {
"opennodes": {
"url": "https://registry.opennodes.io/mcp"
}
}
}{
"servers": {
"opennodes": {
"type": "http",
"url": "https://registry.opennodes.io/mcp"
}
}
}{
"context_servers": {
"opennodes": {
"url": "https://registry.opennodes.io/mcp"
}
}
}{
"mcpServers": {
"opennodes": {
"serverUrl": "https://registry.opennodes.io/mcp"
}
}
}{
"mcpServers": {
"opennodes": {
"url": "https://registry.opennodes.io/mcp"
}
}
}name: OpenNodes
version: 0.0.1
schema: v1
mcpServers:
- name: OpenNodes
type: streamable-http
url: https://registry.opennodes.io/mcp{
"mcpServers": {
"opennodes": {
"httpUrl": "https://registry.opennodes.io/mcp"
}
}
}{
"mcpServers": {
"opennodes": {
"type": "http",
"url": "https://registry.opennodes.io/mcp"
}
}
}{
"mcpServers": {
"opennodes": {
"url": "https://registry.opennodes.io/mcp"
}
}
}goose://extension?url=https%3A%2F%2Fregistry.opennodes.io%2Fmcp&type=streamable_http&id=opennodes&name=OpenNodes
It answered our last check on 21 November 2026 in 44 ms. Over 4 checks since 5 October 2026 it answered 2. Status here comes from Neuronto's own probes, not from the publisher.
7, as returned by the server's own tools/list on 21 November 2026: estimate, find_locations, find_residency, get_node, get_offering, recommend, search_offerings.
It answered our last check on 21 November 2026 in 44 ms. Over 4 checks since 5 October 2026 it answered 2.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://registry.opennodes.io/mcp. The exact JSON is above, with the same setup for Claude Code, Claude Desktop, VS Code, Windsurf, Zed, Gemini CLI and the rest.
It listed its tools to an anonymous client, so reading the tool list needs no key. Individual tools can still ask for credentials in their own inputs.
Publisher: io.github.opennodes-io. Endpoint host: registry.opennodes.io. ARD identifier: urn:air:io.github.opennodes-io:mcp:opennodes.
First indexed 4 October 2026. The tools are exactly what the server returned; nothing on this page
is inferred from its name. Ranking in Neuronto search is never for sale.