plug.directory
Add my server

TabArena

by tabarenaco.huggingface.tabarena/leaderboard

Elo-ranked leaderboards for tabular ML, IID and beyond

Install in oterm Other clients
  • Remote endpoint
  • 4 tools
  • APACHE-2.0

Uptime

Checked  · every day

100%from 4 days of checks

Up without downtime on 4 of 4 days with data.

  • No downtime
  • Degraded
  • Down over 5 minutes
  • No data
  • Not checked yet

Response time

Median 846 ms, slowest 5% 989 ms for the handshake and the tool list, over 30 days.

1000 ms500 ms0

How far each check got

  • Reached the address 100%
  • Completed the handshake 100%
  • Listed its tools 100%

4 tools at the last answer.

Every day
DayResultChecksMedian
3 Nov (today)No downtime1911 ms
2 NovNo downtime1of 1842 ms
1 NovNo downtime1of 1938 ms
31 OctNo downtime2840 ms

A failure counts as down time only when a second check five minutes later fails too. How we check

Connect to the publisher's own endpoint

Add to config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows):

{
  "mcpServers": {
    "leaderboard": {
      "url": "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
    }
  }
}

Step by step for oterm

For agents: the install prompt

Every step above as one Markdown text, for an assistant installing it for you.

# Install the MCP server TabArena (`co.huggingface.tabarena/leaderboard`)

Its page: https://plug.directory/servers/co.huggingface.tabarena/leaderboard

It runs on the publisher's servers; the client connects to its URL.

Use the section for the person's client.

## Claude Desktop

Paste into Settings › Connectors › Add custom connector:

```text
https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse
```

## Claude Code

Run in a terminal:

```sh
claude mcp add --transport sse leaderboard https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse
```

## Cursor, Devin Desktop, LM Studio, Gemini CLI, oterm, ECA

Merge into the place for the person's client:

- Cursor: ~/.cursor/mcp.json
- Devin Desktop: ~/.config/devin/mcp_config.json
- LM Studio: mcp.json (Program › Install › Edit mcp.json)
- Gemini CLI: ~/.gemini/settings.json
- oterm: config.json in oterm's data folder (~/Library/Application Support/oterm on macOS, ~/.local/share/oterm on Linux, %APPDATA%\oterm on Windows)
- ECA: ~/.config/eca/config.json (global) or .eca/config.json in the project

```json
{
  "mcpServers": {
    "leaderboard": {
      "url": "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## VS Code

Merge into .vscode/mcp.json:

```json
{
  "servers": {
    "leaderboard": {
      "type": "sse",
      "url": "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Zed

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Cline, AnythingLLM, Augment Code

Merge into the place for the person's client:

- Cline: cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers)
- AnythingLLM: anythingllm_mcp_servers.json in the AnythingLLM storage plugins directory
- Augment Code: Augment Settings Panel – MCP section – Import from JSON (VS Code extension)

```json
{
  "mcpServers": {
    "leaderboard": {
      "type": "sse",
      "url": "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Continue

Merge into ~/.continue/config.yaml:

```yaml
mcpServers:
  - name: "leaderboard"
    type: sse
    url: "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
```

## Goose

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Codex

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## mcp-client-cli, TUUI, BoltAI, Bob Shell, Kiro, Kiro CLI

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Agenvoy

This server's endpoint speaks only SSE, which this client's documentation does not list among its transports; use another client for it.

## Amp

Merge into ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):

```json
{
  "amp.mcpServers": {
    "leaderboard": {
      "url": "https://tabarena-leaderboard.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

Then restart the client, or reload its MCP servers, and ask it to list the server's tools to confirm it started.

Tools 4

  • leaderboard_get_beyondarena_leaderboard Ranked results for BeyondArena, the companion benchmark that measures the same tabular models (TabPFN, LightGBM, CatBoost, AutoGluon and the rest) where the IID assumption does not hold: random, temporal and grouped splits, sliced by dataset size, feature type and task type, in any combination. Use it whenever the question involves distribution shift, time-ordered rows, leakage-prone grouped splits, or how well a tabular model generalizes beyond a random train/test split. Takes the same kind='models' | 'systems' | 'all' distinction as the TabArena endpoint.
  • leaderboard_get_pareto_frontier The accuracy-versus-time trade-off among tabular models (TabPFN, TabICL, LightGBM, XGBoost, CatBoost, RealMLP, AutoGluon and the rest), measured on TabArena or BeyondArena. Returns the outright best model, the models nothing beats on both accuracy and speed at once (each with its speedup and accuracy gap versus the leader), and the best model that fits a train- or predict-time budget. Prefer it over the leaderboard endpoints for 'which model should I use on my tabular data', 'what is the best tabular model', 'what is a faster alternative to TabPFN', or anything about cost of training or serving: the top of the leaderboard is often orders of magnitude slower than a model just behind it. Returns a summary sentence you can relay.
  • leaderboard_get_tabarena_leaderboard TabArena is a living benchmark for tabular machine learning: predicting a target column from structured, rows-and-columns data (CSV files, spreadsheets, dataframes, database tables). It ranks tabular foundation models (TabPFN, TabICL, TabDPT, Mitra, TabM), gradient-boosted trees (LightGBM, XGBoost, CatBoost), neural networks (RealMLP, ModernNCA) and AutoML systems (AutoGluon) by Elo over 51 curated datasets, with measured training and inference time for each. This endpoint returns the ranked results (Elo, score, ranks, train and predict time) for one subset of the IID benchmark, selected by task type, dataset size, imputation and split protocol. Use it to look up where a named method stands, to compare two of them ("is TabPFN better than LightGBM on small data"), or to report the current state of the art on tabular data. Pass kind='models' (the default) for individual models, kind='systems' for whole AutoML systems such as AutoGluon, or kind='all' for both: a question about the best mo
  • leaderboard_list_leaderboards TabArena is a living benchmark for tabular machine learning: predicting a target column from structured, rows-and-columns data (CSV files, spreadsheets, dataframes, database tables). It ranks tabular foundation models (TabPFN, TabICL, TabDPT, Mitra, TabM), gradient-boosted trees (LightGBM, XGBoost, CatBoost), neural networks (RealMLP, ModernNCA) and AutoML systems (AutoGluon) by Elo over 51 curated datasets, with measured training and inference time for each. This endpoint is the index: it lists the available leaderboards, the valid subset values for every other endpoint, the difference between a model and a whole system, the keys each record carries, and the raw CSV URLs for bulk download. Call it before the get_* endpoints when you are unsure which arguments exist.

From Hugging Face Spaces. Report a problem