LMX Cloud LLM Inference is an MCP server with 9 tools an agent can call, among them chat_completion, extract_pdf, get_balance. It answers at https://mcp.lmxcloud.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 6 of 6 times since 11 September 2026; the last check, on 24 October 2026, succeeded.
OpenAI-compatible LLM MCP (7 tools); chat via balance key or x402 USDC on Base
Description as the publisher wrote it.
| Tool | What it does |
|---|---|
chat_completion | Call LMX Cloud OpenAI-compatible chat completions. Prefers a pre-funded API key (Bearer / api_key); when omitted and x402 is enabled, requires a USDC pay-per-call payment. Optional image_url / images enable vision input (OpenAI content-part Inputs: prompt, image_url, images, model, max_tokens, temperature |
extract_pdf | Extract text and light structure (title, headings, page count) from a PDF via LMX pdf-extract. Provide file_url and/or file_base64. Requires a real LMX API key (same gate as web_search). Inputs: file_url, file_base64, api_key |
get_balance | Fetch USD credit balance for the caller API key. Requires authentication. Inputs: api_key |
get_pricing | Fetch current LMX Cloud per-call pricing catalog. Inputs: api_key |
get_status | Fetch LMX Cloud provider health, fallback chain, and anchoring status. Inputs: api_key |
get_usage | Fetch request and token usage totals for the caller API key. Requires authentication. Inputs: api_key |
list_models | List currently supported LMX model aliases and providers. Inputs: api_key |
quote_price | Estimate USDC cost for a single model call. Uses GET /v1/pricing with model and token params. Inputs: model, max_tokens, prompt_tokens, api_key |
web_search | Real-time web search via LMX (Brave Search passthrough). Fixed per-call USDC price from the caller's API key balance. Returns title/url/snippet results. Inputs: query, max_results, api_key |
Every block below points at https://mcp.lmxcloud.io/mcp. Pick your client and copy.
claude mcp add --transport http lmx-cloud-llm-inference https://mcp.lmxcloud.io/mcp
{
"mcpServers": {
"lmx-cloud-llm-inference": {
"command": "npx",
"args": ["-y", "mcp-remote", "https://mcp.lmxcloud.io/mcp"]
}
}
}{
"mcpServers": {
"lmx-cloud-llm-inference": {
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"servers": {
"lmx-cloud-llm-inference": {
"type": "http",
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"context_servers": {
"lmx-cloud-llm-inference": {
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"mcpServers": {
"lmx-cloud-llm-inference": {
"serverUrl": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"mcpServers": {
"lmx-cloud-llm-inference": {
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}name: LMX Cloud LLM Inference
version: 0.0.1
schema: v1
mcpServers:
- name: LMX Cloud LLM Inference
type: streamable-http
url: https://mcp.lmxcloud.io/mcp{
"mcpServers": {
"lmx-cloud-llm-inference": {
"httpUrl": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"mcpServers": {
"lmx-cloud-llm-inference": {
"type": "http",
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}{
"mcpServers": {
"lmx-cloud-llm-inference": {
"url": "https://mcp.lmxcloud.io/mcp"
}
}
}goose://extension?url=https%3A%2F%2Fmcp.lmxcloud.io%2Fmcp&type=streamable_http&id=lmx-cloud-llm-inference&name=LMX%20Cloud%20LLM%20Inference
It answered our last check on 24 October 2026 in 295 ms. Over 6 checks since 11 September 2026 it answered 6. Status here comes from Neuronto's own probes, not from the publisher.
Yes. It is published by lmxcloud.io and served from mcp.lmxcloud.io, the vendor's own domain: https://mcp.lmxcloud.io/mcp.
9, as returned by the server's own tools/list on 22 September 2026: chat_completion, extract_pdf, get_balance, get_pricing, get_status, get_usage, list_models, quote_price, and 1 more.
It answered our last check on 24 October 2026 in 295 ms. Over 6 checks since 11 September 2026 it answered 6.
Put the server under mcpServers in ~/.cursor/mcp.json with the URL https://mcp.lmxcloud.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.lmxcloud. Endpoint host: mcp.lmxcloud.io. ARD identifier: urn:air:io.lmxcloud:mcp:mcp-server.
First indexed 10 September 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.