INDIVIDUAL MCP TOOL

get_error_reference

Look up what a FlexInference error code means and how to fix it (status, type, full message, and a docs link).

flexinference.comnone authenticationAvailability not checked

LIVE ENDPOINT

https://mcp.flexinference.com/mcp

No auth detected

Connect to this endpoint to inspect the live schema for get_error_reference and invoke it with your own arguments.

{}

Potential side effects detected · medium confidence · heuristic, not a guarantee.

CONNECT WITH APPROVAL

Review this server and its permissions before adding it. Secret placeholders must be set locally.

Codex

~/.codex/config.toml

[mcp_servers.flexinference]
url = "https://mcp.flexinference.com/mcp"
enabled = true

Claude Code

.mcp.json

{
  "mcpServers": {
    "flexinference": {
      "type": "http",
      "url": "https://mcp.flexinference.com/mcp"
    }
  }
}

Claude Desktop

Settings → Connectors → Add custom connector

Name: flexinference
Remote MCP URL: https://mcp.flexinference.com/mcp

Add this remote URL as a custom connector in Claude Desktop. Availability depends on the user plan and workspace policy.

Cursor

.cursor/mcp.json

{
  "mcpServers": {
    "flexinference": {
      "url": "https://mcp.flexinference.com/mcp"
    }
  }
}

Visual Studio Code

.vscode/mcp.json

Add to Visual Studio Code

{
  "servers": {
    "flexinference": {
      "type": "http",
      "url": "https://mcp.flexinference.com/mcp"
    }
  }
}

Generic MCP

Client-specific MCP configuration

{
  "name": "flexinference",
  "transport": "streamable-http",
  "url": "https://mcp.flexinference.com/mcp"
}

MCP Inspector

Run the official MCP Inspector locally and enter the indexed Streamable HTTP endpoint.

search_docs — Keyword search over the FlexInference documentation (quickstart, authentication, deadline routing, supported models, SDK usage, and the four API endpoints).
run_flex_demo — Run the homepage demo programmatically: one prompt is sent twice on the same model, once through the flex race and once as a plain default request, and the answer carries both outputs plus the true billed comparison (cost per lane and time to first token).
compare_usage_export — Paste the text of a usage CSV export (OpenAI platform, Anthropic console, or OpenRouter) and get back what those rows would have cost through FlexInference: total cost, savings at each model's measured realized rate, and the time-to-first-token trade, per model and in total.
get_usage_summary — This organization's flex-routed usage and savings: total requests, what was billed, what standard-tier pricing would have cost, and the difference, broken down by model.
list_api_keys — List this organization's FlexInference API keys: name, last 4 characters, creation time, and a truncated id (never the full key - it is only ever shown once, at creation).
get_byok_status — Check which providers (OpenAI, Gemini, Anthropic, Bedrock, Vertex) have a BYOK key configured for this organization.
get_billing_status — Read this organization's billing status.