# LLM Configurator

> LLM Configurator is a free MCP server for ai models and llm tooling. Check whether a GPU or Mac can run a local LLM: memory fit, speed estimate, model specs. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** Web, API
- **Website:** https://llmconfigurator.com/en/mcp
- **Repository:** https://github.com/aikuba/llmconfigurator.com
- **MCP Registry name:** io.github.aikuba/llmconfigurator
- **Categories:** AI models and LLM tooling
- **Listing updated:** 2026-11-16

## About

Instead of answering hardware questions from memory, an assistant can call this read-only server, which uses the same calculations as the LLM Configurator website. Ask whether an RTX 4090 can run Qwen3 32B and it reports the memory arithmetic and a speed range labelled as an estimate. Ask what a 64 GB Mac Studio can run and it returns a ranked list with a fit verdict for each model.

Its examples show how it handles mixture-of-experts models: Llama 4 Scout is said to need about 65.8 GB because every expert must be resident, even though only 17B parameters are active per token. It can also compare GPUs on memory, bandwidth and any dated price on record, with an unknown price returned as null, and give a model's total and active parameters, weight size at each quantisation, context length and licence. The page names five tools but not their identifiers.

The endpoint is free, needs no sign-in and is hosted on Supabase. The privacy section says it keeps only anonymous per-request records for 30 days, such as the UTC day, which tool was called, a rough client type, success and latency, plus a daily-rotating hash of the network address. Tool arguments and search text are never stored.

## Features

- **Fit check:** Says whether a model fits a given GPU or Mac and shows the memory arithmetic.
- **Speed estimates:** Gives tokens-per-second ranges labelled as estimates.
- **Hardware ranking:** Lists which models a machine can run, with a fit verdict for each.
- **GPU comparison:** Compares memory, bandwidth and any dated price on record.
- **Model specs:** Returns parameters, quantised sizes, context length and licence.
- **Anonymous logging:** Stores no questions or addresses; records are deleted after 30 days.

## Good for

- Check if a 24 GB card can run a mixture-of-experts model
- Pick models for a 64 GB Mac Studio
- Compare two GPUs for local inference
- Look up a model's licence and Q4 size

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http llmconfigurator https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp
```

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

```sh
https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp
```

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

```json
{
  "mcpServers": {
    "llmconfigurator": {
      "url": "https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp"
    }
  }
}
```

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

```json
{
  "servers": {
    "llmconfigurator": {
      "type": "http",
      "url": "https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "llmconfigurator": {
      "serverUrl": "https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp"
    }
  }
}
```

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

```json
{
  "mcpServers": {
    "llmconfigurator": {
      "type": "http",
      "url": "https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp"
    }
  }
}
```

## FAQ

### Is it free and does it need sign-in?

Yes, it is free with no authentication, and it is read-only.

### Are speed numbers exact?

No. The site labels speed ranges as estimates.

### What does it log?

Anonymous request records such as the day, tool used and latency, deleted after 30 days. It does not store what you asked.

### Does it work in ChatGPT?

The site describes setup through ChatGPT developer mode, which is not available on every account.

### Is LLM Configurator free?

Yes. LLM Configurator is free to use.

### How do I add LLM Configurator to Claude Code?

Run this command in your terminal: claude mcp add --transport http llmconfigurator https://elgadmisijrfowuujcmu.supabase.co/functions/v1/mcp

### What are the best alternatives to LLM Configurator?

The closest alternatives to LLM Configurator are Claude Code Ultimate Guide, signafAIde, Toolkien, Lugh and HypePaper.

## 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/llm-configurator
