# InferCPI

> InferCPI is a free MCP server for analytics and data. Query AI inference price indexes, model prices, speed and cost per task. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** API, CLI
- **Website:** https://infercpi.com/api.html
- **Repository:** https://github.com/infer-cpi/api
- **MCP Registry name:** com.infercpi/api
- **Categories:** analytics and data, AI models and LLM tooling
- **Listing updated:** 2026-11-14

## About

Everything here comes from InferCPI's own index of what AI inference costs. Its series cover list prices, what buyers actually pay (usage-weighted), the price of a fixed level of capability, open versus closed weights, third-party hosts, GPU rental, self-hosting break-even and dollars per completed task.

The MCP server wraps the same static JSON files as the public API in ten read-only tools, from the latest value of every series to a weekly basket of priced models, price-change events since September 2023, daily time-to-first-token and throughput measurements, and cost-per-task figures. Inputs include LiteLLM and genai-prices, the Epoch AI Capabilities Index, OpenRouter rankings and SEC filings; speed and cost per task are the site's own measurements.

A hosted endpoint needs no key, or you can download the single Python file and run it yourself. Data is licensed CC BY 4.0 and the code MIT.

## Features

- **Price index series:** Chained, quality-adjusted and usage-weighted indexes of inference prices.
- **Weekly model basket:** Input, output and blended dollars per million tokens for each priced model.
- **Price-change events:** Launches, cuts, increases and removals from labs' price lists since Sep 2023.
- **Speed measurements:** Daily time to first token and output tokens per second by model and route.
- **Self-hosting comparison:** Compares rented or owned GPUs with API prices, including break-even.
- **Cost per task:** Dollars and tokens per completed task, with accuracy and reasoning share.

## Good for

- Compare what a model costs across providers before choosing one
- Track whether open-weight models are getting cheaper than closed ones
- Decide if renting GPUs beats an API at your volume
- Add current price data to an analyst or finance agent
- Cite inference price trends with a CC BY source

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http api https://infercpi.com/mcp
```

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

```sh
https://infercpi.com/mcp
```

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

```json
{
  "mcpServers": {
    "api": {
      "url": "https://infercpi.com/mcp"
    }
  }
}
```

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

```json
{
  "servers": {
    "api": {
      "type": "http",
      "url": "https://infercpi.com/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "api": {
      "serverUrl": "https://infercpi.com/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "api": {
      "type": "http",
      "url": "https://infercpi.com/mcp"
    }
  }
}
```

## Tools

- `index_latest`
- `index_series`
- `index_catalogue`
- `basket`
- `model_price`
- `events`
- `speed`
- `open_vs_closed`
- `self_hosting`
- `cost_per_task`

## FAQ

### Is a key required?

No. The hosted endpoint and the data files need no key.

### What licence applies to the data?

CC BY 4.0, citing InferCPI, infercpi.com. The code is MIT.

### Can I self-host the server?

Yes. It is a single Python file that needs the mcp package, and INFERCPI_BASE can point it at another data source.

### Where do the numbers come from?

Public sources such as LiteLLM, genai-prices, Epoch AI, OpenRouter rankings and SEC filings, plus the site's own speed and cost-per-task runs.

### Is InferCPI free?

Yes. InferCPI is free to use.

### How do I add InferCPI to Claude Code?

Run this command in your terminal: claude mcp add --transport http api https://infercpi.com/mcp

### What are the best alternatives to InferCPI?

The closest alternatives to InferCPI are Agent-Native Analytics, Claude Code Ultimate Guide, Google Ads + Meta Ads + SEO/GEO MCP, Tapetide - Indian Stock Market MCP and World Monitor.

## Alternatives

- [Agent-Native Analytics](https://ai-powered.com/mcp/agent-native-analytics): Agent-Native Analytics is an open-source MCP server for analytics and data.
- [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.
- [Google Ads + Meta Ads + SEO/GEO MCP](https://ai-powered.com/mcp/google-ads-meta-ads-seo-geo-mcp): Google Ads + Meta Ads + SEO/GEO MCP is a freemium MCP server for marketing and seo.
- [Tapetide - Indian Stock Market MCP](https://ai-powered.com/mcp/tapetide-indian-stock-market-mcp): Tapetide - Indian Stock Market MCP is a freemium MCP server for finance and payments.
- [World Monitor](https://ai-powered.com/mcp/world-monitor): World Monitor is a freemium MCP server for analytics and data.
- [CompanyData](https://ai-powered.com/mcp/companydata): CompanyData is a paid MCP server for analytics and data.
- [ChartAnt](https://ai-powered.com/mcp/chartant): ChartAnt is a freemium MCP server for design and media.
- [Quantivex Research](https://ai-powered.com/mcp/quantivex-research): Quantivex Research is a free MCP server for finance and payments.

---
Source: https://ai-powered.com/mcp/infercpi
