# Openmark MCP server

Source: https://glimind.com/server/mcp-registry%2Fai.openmark%2Fopenmark
Registry id: `mcp-registry/ai.openmark/openmark`
Status: live (checked 2026-10-27T04:09:56.290Z)

## Measurements (30 days)

- Uptime: 100%
- Checks: 2153
- Outages (90 days): 0

## Install

```json
{
  "mcpServers": {
    "openmark-mcp-server": {
      "type": "http",
      "url": "https://openmark.ai/mcp"
    }
  }
}
```

Claude Code:

```sh
claude mcp add --transport http openmark-mcp-server https://openmark.ai/mcp
```

## Tools (9)

- `estimate_cost`: Credits a run_benchmark call would use, before spending anything: the expected total, the split per model, the balance needed to start, the worst case, and the 
- `get_account`: The user's plan, credit balance and limits, and what a credit pack, Pro or Expert would unlock. Use it before large runs, or when a run is refused for credits o
- `get_results`: Status of a benchmark job, and once completed the ranked results (score, stability, cost per run, latency), the tests each model missed with its answer, a recom
- `get_task_format`: How to write an OpenMark task: the YAML structure, all 18 scoring modes, and tips. Call this before writing a task. Pass a topic for the full detail on one subj
- `list_models`: Current OpenMark model ids with list prices ($ per 1M tokens), as text and as a structured list. Use these ids in run_benchmark. Append @low, @medium or @high t
- `prepare_attachments`: Get an upload_uid and a one-hour upload link (no API key needed) for images, PDFs or text documents, with the exact curl command. Upload the files, reference th
- `run_benchmark`: Save a task (YAML, see get_task_format) to the user's OpenMark account and run it on the given models with real API calls. Returns a job_id and the expected cos
- `validate_task`: Check a task YAML the way OpenMark will before running it: structure, scoring settings, and the test limit of the user's plan. Free, saves nothing. Returns the 
- `watch_task`: Watch the task of a completed job: when new or cheaper models ship, OpenMark re-tests up to 4 of them next to the user's pick and emails the user only when one 

Measured by Glimind (https://glimind.com): MCP handshake + tool list only, no tool is ever called. CC BY 4.0.