# BenchLeader

> BenchLeader is a free MCP server for ai models and llm tooling. Free read-only MCP server for the daily LLM leaderboard, benchmarks and model comparisons. It runs as a hosted remote server, so there is nothing to install locally.

- **Kind:** MCP server
- **Pricing:** Free
- **Platforms:** API, Web
- **Website:** https://www.benchleader.com/mcp
- **MCP Registry name:** com.benchleader/benchleader
- **Categories:** AI models and LLM tooling
- **Listing updated:** 2026-11-15

## About

Assistants can answer questions about which model leads, how two compare or what a benchmark measures from BenchLeader's figures instead of memory. The server is free, read-only and needs no key or OAuth, and it speaks Streamable HTTP with stateless JSON responses. The page states data as of 14 Nov 2026, Index v1.4, refreshed with each daily update.

The BenchLeader Index is a composite of independent public benchmarks, with one row per model using its best reasoning-effort configuration. Model profiles add category scores, list price, speed, latency and context window. Comparisons claim a lead only when it falls outside the combined margin of error.

BenchLeader's own computed figures are licensed CC BY 4.0 with attribution, while list prices and publisher benchmark results stay under their owners' terms. Paging limits apply, for example 100 rows per leaderboard call, and the server is not meant as a bulk download.

## Features

- **Daily leaderboard:** Ranks models by the Index or by one category such as coding or math.
- **Model search:** Finds models by name, alias, id or developer.
- **Head-to-head comparison:** Gives a verdict with price and speed ratios and category wins.
- **Benchmark catalogue:** Explains what each benchmark measures and its weight in the Index.
- **Source links:** Every result carries BenchLeader and publisher URLs.
- **No key needed:** Free, read-only and open to any MCP client.

## Good for

- Choose a model for a coding workload by category score
- Compare two models on price and speed
- Look up what a benchmark measures before trusting a score
- Have an assistant cite current rankings in a report

## Install

### Claude Code (Terminal)

```sh
claude mcp add --transport http benchleader https://www.benchleader.com/mcp
```

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

```sh
https://www.benchleader.com/mcp
```

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

```json
{
  "mcpServers": {
    "benchleader": {
      "url": "https://www.benchleader.com/mcp"
    }
  }
}
```

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

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

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

```json
{
  "mcpServers": {
    "benchleader": {
      "serverUrl": "https://www.benchleader.com/mcp"
    }
  }
}
```

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

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

## Tools

- `get_leaderboard`: Returns the BenchLeader Index leaderboard, optionally by category.
- `search_models`: Finds models by name, alias, id or developer.
- `get_model`: Returns one model's profile with scores, price and speed.
- `compare_models`: Compares two models head to head.
- `list_benchmarks`: Lists the benchmarks BenchLeader reads.
- `get_benchmark`: Returns one benchmark and its top models.
- `about_benchleader`: Explains the Index method, cadence and licence.

## FAQ

### Do I need an API key?

No. The server is free and needs no key or OAuth.

### Can it change data?

No, it is read-only.

### How do I cite the figures?

Credit BenchLeader with a link to benchleader.com; its own figures are CC BY 4.0.

### Which clients can connect?

The page gives setup for Claude Desktop, Claude Code, Cursor, VS Code, Windsurf and any Streamable HTTP client.

### Is BenchLeader free?

Yes. BenchLeader is free to use.

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

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

### What are the best alternatives to BenchLeader?

The closest alternatives to BenchLeader 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/benchleader
