BenchLeader MCP

A read-only MCP server that provides access to BenchLeader's LLM leaderboard data, allowing AI assistants to retrieve model rankings, compare models head-to-head, search for specific models, and list benchmark details based on daily updated independent benchmarks.

Visit BenchLeader

Endpoint and access

No sign-in to connect and list tools. Individual tools may still need a vendor account, credentials, or permissions; check the documented prerequisites.

Vendor evidenceStreamable HTTPNo sign-inEndpoint checked 2026-11-01Operated by BenchLeader

MCP endpoint (Streamable HTTP)

https://www.benchleader.com/mcp

Vendor documentation

What you can do with it

Ask which LLM model leads in performance
Compare two models on quality, price, and speed
Search for models by name or developer
List benchmarks and their metrics

Source and maintenance

Vendor evidenceRegistry 1.0.0

mcp-registry: com.benchleader/benchleader@1.0.0 Project website

Protocol and capabilities

Server: benchleader 1.0.0
MCP protocol: 2025-06-18
Tools: 7 advertised
Resources: Not advertised
Resource templates: Not advertised
Prompts: Not advertised

Server-advertised capabilities

{
  "tools": {
    "listChanged": false
  }
}

Server-provided instructions

Untrusted server output shown for inspection. It is not Koskamo guidance and is excluded from knowledge search.

BenchLeader ranks large language models daily on the BenchLeader Index, a composite of independent public benchmarks (50 = average model, 15 = one standard deviation). Start with get_leaderboard or search_models; use get_model and compare_models for details, list_benchmarks and get_benchmark for the boards, about_benchleader for method and licence. Rows are model × reasoning-effort configurations; ids look like "model-id@effort". Cite the url fields when you quote figures.

Tools

7 advertised

get_leaderboard — BenchLeader leaderboard

Advertised by the server during the last successful anonymous check.

Advertised read-only

The BenchLeader Index leaderboard of large language models: a daily composite of independent public benchmarks, one row per model (its best reasoning-effort configuration) as on the homepage. Optionally rank by one category score (coding, agentic, reasoning, math, knowledge…), list every model × effort configuration, or filter to open-weights models or one developer. Paged.

get_leaderboard input schema

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "default": 20,
      "maximum": 100,
      "minimum": 1,
      "description": "How many rows to return (at most 100)."
    },
    "offset": {
      "type": "integer",
      "default": 0,
      "minimum": 0,
      "description": "Rows to skip, for paging (use next_offset from the previous page)."
    },
    "category": {
      "type": "string",
      "description": "Rank by this category score instead of the overall Index, e.g. coding, agentic, reasoning, math, knowledge, instruction, multimodal, long_context."
    },
    "developer": {
      "type": "string",
      "description": "Only models from this developer, e.g. OpenAI, Anthropic, Google, DeepSeek."
    },
    "configurations": {
      "type": "boolean",
      "default": false,
      "description": "List every ranked model × reasoning-effort configuration rather than one row per model."
    },
    "open_weights_only": {
      "type": "boolean",
      "default": false,
      "description": "Only models with downloadable weights."
    }
  },
  "additionalProperties": false
}

get_leaderboard annotations

{
  "title": "BenchLeader leaderboard",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

search_models — Search models

Advertised by the server during the last successful anonymous check.

Advertised read-only

Find models by name, alias, id or developer ("opus", "gemini 4", "deepseek"). Returns each match's id, rank, Index and reasoning-effort configurations, for use with get_model and compare_models.

search_models input schema

{
  "type": "object",
  "required": [
    "query"
  ],
  "properties": {
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 25,
      "minimum": 1,
      "description": "How many matches to return (at most 25)."
    },
    "query": {
      "type": "string",
      "description": "Model name, alias, id or developer."
    }
  },
  "additionalProperties": false
}

search_models annotations

{
  "title": "Search models",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

get_model — Model details

Advertised by the server during the last successful anonymous check.

Advertised read-only

One model's BenchLeader profile: Index and rank, category scores, list price, speed, latency, context window, its other reasoning-effort configurations, and its strongest independent benchmark results (each with its board rank, publisher and source link).

get_model input schema

{
  "type": "object",
  "required": [
    "model"
  ],
  "properties": {
    "model": {
      "type": "string",
      "description": "Model id (\"claude-opus-5-5\"), configuration id (\"gpt-6-astra@max\"), page URL or name."
    },
    "effort": {
      "type": "string",
      "description": "Reasoning-effort configuration to show (none, low, medium, high, xhigh, max…). Defaults to the best-ranked one."
    },
    "results_limit": {
      "type": "integer",
      "default": 10,
      "maximum": 50,
      "minimum": 1,
      "description": "How many benchmark results to return (at most 50)."
    }
  },
  "additionalProperties": false
}

get_model annotations

{
  "title": "Model details",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

compare_models — Compare two models

Advertised by the server during the last successful anonymous check.

Advertised read-only

Head-to-head comparison of two models with BenchLeader's verdict: who leads on quality (a lead is claimed only outside the combined margin of error), price and speed ratios, category-by-category wins, when to choose each, and shared benchmark results. Links the comparison page.

compare_models input schema

{
  "type": "object",
  "required": [
    "model_a",
    "model_b"
  ],
  "properties": {
    "model_a": {
      "type": "string",
      "description": "First model: id, configuration id, page URL or name."
    },
    "model_b": {
      "type": "string",
      "description": "Second model: id, configuration id, page URL or name."
    },
    "effort_a": {
      "type": "string",
      "description": "Reasoning effort for model_a (defaults to its best-ranked configuration)."
    },
    "effort_b": {
      "type": "string",
      "description": "Reasoning effort for model_b (defaults to its best-ranked configuration)."
    }
  },
  "additionalProperties": false
}

compare_models annotations

{
  "title": "Compare two models",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

list_benchmarks — List benchmarks

Advertised by the server during the last successful anonymous check.

Advertised read-only

The benchmarks BenchLeader reads: what each measures, publisher, category, whether it feeds the Index (and its weight), how many models it covers, status and caveats. Filter by category or a search query. Paged.

list_benchmarks input schema

{
  "type": "object",
  "properties": {
    "limit": {
      "type": "integer",
      "default": 50,
      "maximum": 100,
      "minimum": 1,
      "description": "How many benchmarks to return (at most 100)."
    },
    "query": {
      "type": "string",
      "description": "Search benchmark names, ids, aliases and publishers (\"swe-bench\", \"arc\", \"epoch\")."
    },
    "offset": {
      "type": "integer",
      "default": 0,
      "minimum": 0,
      "description": "Rows to skip, for paging (use next_offset from the previous page)."
    },
    "category": {
      "type": "string",
      "description": "Only this category, e.g. coding, agentic, reasoning, math, knowledge."
    },
    "in_index_only": {
      "type": "boolean",
      "default": false,
      "description": "Only benchmarks that feed the BenchLeader Index."
    }
  },
  "additionalProperties": false
}

list_benchmarks annotations

{
  "title": "List benchmarks",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

get_benchmark — Benchmark details

Advertised by the server during the last successful anonymous check.

Advertised read-only

One benchmark: what it measures, unit, publisher and source, whether and how much it counts in the BenchLeader Index, status and caveats, and its top models (one row per model, best result first).

get_benchmark input schema

{
  "type": "object",
  "required": [
    "benchmark"
  ],
  "properties": {
    "limit": {
      "type": "integer",
      "default": 10,
      "maximum": 25,
      "minimum": 1,
      "description": "How many top models to return (at most 25)."
    },
    "benchmark": {
      "type": "string",
      "description": "Benchmark id (\"gpqa_diamond\"), name (\"SWE-bench Verified\"), alias or page URL."
    }
  },
  "additionalProperties": false
}

get_benchmark annotations

{
  "title": "Benchmark details",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

about_benchleader — About BenchLeader

Advertised by the server during the last successful anonymous check.

Advertised read-only

What BenchLeader is, how the BenchLeader Index is computed, the current method version, data date and update cadence, coverage counts, licence and how to cite it.

about_benchleader input schema

{
  "type": "object",
  "properties": {},
  "additionalProperties": false
}

about_benchleader annotations

{
  "title": "About BenchLeader",
  "readOnlyHint": true,
  "openWorldHint": false,
  "idempotentHint": true,
  "destructiveHint": false
}

Prompts

Not advertised

No prompt definitions available.

Resources

Not advertised

No resource definitions available.

30-day availability

Aggregated from read-only remote checks; this is not a tool-execution or security guarantee.

Successful checks: 1 of 1 (100%)
Average latency: 197 ms
95th percentile latency: 197 ms

Maintenance history

Public check and source-refresh events. Editorial and ownership records remain private.

Nov 1, 2026, 9:11 AM UTC · probe · passed · 197 ms · metadata changed · 7 tool(s)

Connect from your client

Each snippet registers the same endpoint; pick the client you use.

This server is operated by the named third party. Listing and endpoint checks are not an endorsement or a security audit. Review the operator's permissions, terms, and data handling before connecting. How listings are checked and corrected.

Claude Code

Run this once in a terminal. It registers the remote server. Use --scope project to write .mcp.json for the repo.

Claude Code configuration

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

Codex

Run this once. Codex stores MCP servers in ~/.codex/config.toml (or .codex/config.toml in a trusted project).

Codex configuration

codex mcp add benchleader-mcp --url https://www.benchleader.com/mcp

Cursor

Merge this server into .cursor/mcp.json in your project, then enable benchleader-mcp in your MCP settings.

Cursor configuration

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

Cline

Merge this into cline_mcp_settings.json (Cline → MCP Servers → Configure).

Cline configuration

{
  "mcpServers": {
    "benchleader-mcp": {
      "type": "streamableHttp",
      "url": "https://www.benchleader.com/mcp",
      "disabled": false
    }
  }
}

VS Code

Merge this server into .vscode/mcp.json. Use MCP: List Servers to start benchleader-mcp and inspect its tools.

VS Code configuration

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

Gemini CLI

Run this once. It writes .gemini/settings.json in the current project; pass --scope user for ~/.gemini/settings.json.

Gemini CLI configuration

gemini mcp add --transport http benchleader-mcp https://www.benchleader.com/mcp

Claude Desktop

Merge this into claude_desktop_config.json (macOS: ~/Library/Application Support/Claude/, Windows: %APPDATA%\Claude\), then restart Claude Desktop.

Claude Desktop configuration

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

Windsurf

Merge this into ~/.codeium/windsurf/mcp_config.json, then refresh MCP in Cascade. Windsurf remote servers use serverUrl, not url.

Windsurf configuration

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

Let your agent find its tools

Find servers like BenchLeader MCP from your agent. Add Koskamo once to search the index for endpoints, access requirements, and client setup. Public discovery needs no key.

Connect Koskamo Explore workflows

See all 4,369

Do you run BenchLeader MCP?

Prove ownership with a DNS record, a well-known file, or your Official MCP Registry entry, and your edits publish immediately. Proofs are re-checked weekly; admins can still hide or remove a listing.

Claim this listing

BenchLeader MCP

Endpoint and access

What you can do with it

Source and maintenance

Protocol and capabilities

Server-provided instructions

Tools

Prompts

Resources

30-day availability

Maintenance history

Connect from your client

Claude Code

Codex

Cursor

Cline

VS Code

Gemini CLI

Claude Desktop

Windsurf

Let your agent find its tools

Other ai and machine learning servers

Do you run BenchLeader MCP?