# DeepInfra MCP by usefulapi

> DeepInfra MCP by usefulapi is a freemium MCP server for ai models and llm tooling. Hosted MCP server giving AI clients access to DeepInfra usage, models and deployments. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Freemium
- **Starting price:** $9 a month
- **Platforms:** API, Web, vscode
- **Website:** https://usefulapi.io/deepinfra/
- **Repository:** https://github.com/m190/usefulapi-mcp/tree/HEAD/servers/deepinfra
- **MCP Registry name:** io.usefulapi/deepinfra
- **Categories:** AI models and LLM tooling
- **Listing updated:** 2026-11-18

## About

The README and product page describe this as an independent, hosted connector that lets Claude, ChatGPT, Cursor and other MCP clients query a DeepInfra account. Instead of running anything locally, you add one remote URL and sign in with your DeepInfra API key through an OAuth login page.

The server exposes 22 tools. Most are read-only and cover spend and token usage, invoices, the model catalog with pricing and context length, dedicated deployment status and stats, GPU availability and request logs. Two tools can start or stop a dedicated deployment, which affects GPU charges, so clients are expected to ask for confirmation before running them.

The page says it is aimed at developers and ML engineers who run models or dedicated GPU deployments on DeepInfra and want to check costs or deployment health from within an assistant. It is not an official DeepInfra product, and the page states that it is operated by usefulapi.

## Features

- **Hosted remote connection:** Connects through a single Streamable HTTP URL with OAuth login, so no local install is needed.
- **Spend and token reporting:** Reads billing-period spend, per-model token usage, GPU rental usage and invoices.
- **Model catalog lookup:** Lists available models and reads pricing, context length and capabilities for each.
- **Deployment monitoring:** Shows dedicated deployment config, status, throughput, latency, replicas and request logs.
- **Start and stop deployments:** Two write tools pause or resume dedicated deployments to control GPU charges.

## Good for

- Reviewing DeepInfra spend and token usage for a billing month from a chat assistant
- Comparing model prices and context lengths before choosing one for a project
- Checking whether a dedicated deployment is healthy or idle
- Stopping deployments that receive no traffic to avoid GPU charges
- Planning a new deployment by checking which GPUs can currently be provisioned

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http deepinfra https://deepinfra.usefulapi.io/mcp
```

### Claude Desktop (Settings → Connectors → Add custom connector)

```sh
https://deepinfra.usefulapi.io/mcp
```

### Cursor (~/.cursor/mcp.json)

```json
{
  "mcpServers": {
    "deepinfra": {
      "url": "https://deepinfra.usefulapi.io/mcp"
    }
  }
}
```

### VS Code (.vscode/mcp.json)

```json
{
  "servers": {
    "deepinfra": {
      "type": "http",
      "url": "https://deepinfra.usefulapi.io/mcp"
    }
  }
}
```

### Windsurf (~/.codeium/windsurf/mcp_config.json)

```json
{
  "mcpServers": {
    "deepinfra": {
      "serverUrl": "https://deepinfra.usefulapi.io/mcp"
    }
  }
}
```

### Other clients (your client's MCP settings (most use this shape))

```json
{
  "mcpServers": {
    "deepinfra": {
      "type": "http",
      "url": "https://deepinfra.usefulapi.io/mcp"
    }
  }
}
```

## Tools

- `deepinfra_get_account`: Get the current DeepInfra account details.
- `deepinfra_get_rate_limit`: Get the account's current rate limits.
- `deepinfra_list_api_tokens`: List the account's API tokens (metadata only).
- `deepinfra_get_usage`: Get spend and usage for a billing period.
- `deepinfra_get_usage_tokens`: Get per-model token usage for a period.
- `deepinfra_get_usage_rent`: Get GPU rental usage for a time range.
- `deepinfra_list_invoices`: List billing invoices for the account.
- `deepinfra_list_deployments`: List the account's dedicated deployments.
- `deepinfra_get_deployment`: Get one dedicated deployment's config and status.
- `deepinfra_get_deployment_stats`: Get time-series stats for a dedicated deployment.
- `deepinfra_get_gpu_availability`: Get GPU availability for LLM deployments.
- `deepinfra_list_models`: List the DeepInfra model catalog.
- `deepinfra_get_model`: Get one model's catalog entry.
- `deepinfra_get_hardware`: Get hardware options for a given model.
- `deepinfra_query_logs`: Query inference request logs for a deployment.
- `deepinfra_get_live_metrics`: Get global live inference metrics for the account.
- `deepinfra_start_deployment`: Start (resume) a dedicated deployment.
- `deepinfra_stop_deployment`: Stop (pause) a running dedicated deployment.
- `deepinfra_usage_status`: Report current usage against the plan limit.
- `deepinfra_request_feature`: Send a request for a feature the tools cannot yet provide.
- `deepinfra_upgrade`: Get a paid Pro or Pro+ plan checkout link.
- `deepinfra_cancel_subscription`: Cancel the Pro subscription at the end of the paid period.

## FAQ

### Is this an official DeepInfra product?

No. The product page says usefulapi is an independent service that is not affiliated with or endorsed by DeepInfra.

### What do I need to connect?

The login page asks for your DeepInfra API key, which you create in DeepInfra under Settings and API.

### Which AI clients are supported?

Any client that supports remote MCP servers over Streamable HTTP with OAuth login, including Claude, ChatGPT, Claude Code, Codex, Cursor, VS Code and Windsurf.

### Can the AI change data in my account?

Two tools can start or stop dedicated deployments. The page says most MCP clients ask for approval before a tool call runs.

### Is DeepInfra MCP by usefulapi free?

DeepInfra MCP by usefulapi has a free plan, with paid plans for more usage or features. Paid plans start at $9 a month.

### How do I add DeepInfra MCP by usefulapi to Claude Code?

Run this command in your terminal: claude mcp add --transport http deepinfra https://deepinfra.usefulapi.io/mcp

### What are the best alternatives to DeepInfra MCP by usefulapi?

The closest alternatives to DeepInfra MCP by usefulapi are Claude Code Ultimate Guide, signafAIde, Toolkien, Lugh and HypePaper.

## Where this comes from

- Description: written from the project's README, the project's website and its MCP Registry entry.
- Install steps: from the official MCP Registry entry.

## Alternatives

- [Claude Code Ultimate Guide](https://ai-powered.com/mcp/claude-code-ultimate-guide): Claude Code Ultimate Guide is an open-source MCP server for ai models and llm tooling.
- [signafAIde](https://ai-powered.com/mcp/signafaide): signafAIde is a freemium MCP server for security.
- [Toolkien](https://ai-powered.com/mcp/toolkien): Toolkien is an open-source MCP server for ai models and llm tooling.
- [Lugh](https://ai-powered.com/mcp/lugh): Lugh is an open-source MCP server for ai models and llm tooling.
- [HypePaper](https://ai-powered.com/mcp/hypepaper): HypePaper is a freemium MCP server for ai models and llm tooling.
- [Agent Data Tools: Compact (find, describe, run)](https://ai-powered.com/mcp/agent-data-tools-compact-find-describe-run): Agent Data Tools: Compact (find, describe, run) is a freemium MCP server for developer tools.
- [Angel Agent Guard: pre-purchase check + prompt-injection scan](https://ai-powered.com/mcp/angel-agent-guard-pre-purchase-check-prompt-injection-scan): Angel Agent Guard: pre-purchase check + prompt-injection scan is a freemium MCP server for ai models and llm tooling.
- [Mizan Calibration](https://ai-powered.com/mcp/mizan-calibration): Mizan Calibration is a free MCP server for ai models and llm tooling.

---
Source: https://ai-powered.com/mcp/deepinfra-mcp-by-usefulapi
