NeuroSAM3

by mmrechco.huggingface.mmrech/neurosam3

Agentic neuroimaging — SAM 3, MedSAM, BiomedCLIP, LLM agent

Install in ECA Other clients

Remote endpoint
11 tools
APACHE-2.0

Uptime

Checked 2 Nov, 20:33 UTC · every day

100%from 4 days of checks

Up without downtime on 3 of 3 days with data.

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

Response time

Median 1080 ms, slowest 5% 1238 ms for the handshake and the tool list, over 30 days.

1500 ms750 ms0

How far each check got

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

11 tools at the last answer.

Every day

Day	Result	Checks	Median
3 Nov (today)	No verdict yet	0	–
2 Nov	No downtime	1of 1	1097 ms
1 Nov	No downtime	1of 1	1231 ms
31 Oct	No downtime	2	1072 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/eca/config.json (global) or .eca/config.json in the project:

{
  "mcpServers": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

Step by step for ECA

Add to ~/.config/zed/settings.json:

{
  "context_servers": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

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

Step by step for Zed

Add to cline_mcp_settings.json (MCP Servers › Configure › Configure MCP Servers):

{
  "mcpServers": {
    "neurosam3": {
      "type": "sse",
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

Step by step for Cline

Add to ~/.config/goose/config.yaml:

extensions:
  "neurosam3":
    type: streamable_http
    name: "neurosam3"
    enabled: true
    uri: "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    headers: {}
    env_keys: []
    envs: {}
    timeout: 300

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

Step by step for Goose

Run in a terminal:

codex mcp add neurosam3 --url https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse

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

Step by step for Codex

Add to ~/.llm/config.json:

{
  "mcpServers": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

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

Step by step for mcp-client-cli

Add to ~/.config/agenvoy/mcp.json:

{
  "servers": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

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

Step by step for Agenvoy

Add to ~/.config/amp/settings.json (user) or .amp/settings.json (workspace):

{
  "amp.mcpServers": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}

Step by step for Amp

For agents: the install prompt

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

# Install the MCP server NeuroSAM3 (`co.huggingface.mmrech/neurosam3`)

Its page: https://plug.directory/servers/co.huggingface.mmrech/neurosam3

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://mmrech-neurosam3.hf.space/gradio_api/mcp/sse
```

## Claude Code

Run in a terminal:

```sh
claude mcp add --transport sse neurosam3 https://mmrech-neurosam3.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": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## VS Code

Merge into .vscode/mcp.json:

```json
{
  "servers": {
    "neurosam3": {
      "type": "sse",
      "url": "https://mmrech-neurosam3.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": {
    "neurosam3": {
      "type": "sse",
      "url": "https://mmrech-neurosam3.hf.space/gradio_api/mcp/sse"
    }
  }
}
```

## Continue

Merge into ~/.continue/config.yaml:

```yaml
mcpServers:
  - name: "neurosam3"
    type: sse
    url: "https://mmrech-neurosam3.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": {
    "neurosam3": {
      "url": "https://mmrech-neurosam3.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 11

NeuroSAM3_agent_chat Agent Chat: send a natural-language request about the uploaded image. The agent (LLM via HF Inference Providers) plans tool calls — BiomedCLIP classification, SAM 3 / MedSAM segmentation, ROI measurement, report — and returns a synthesized answer plus overlay images. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_automatic_masks Automatic mask generation: a grid of point prompts through the SAM 3 tracker, NMS by IoU. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_clinical_pipeline Clinical Pipeline: classify -> segment -> measure -> report in one click. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_compare_ground_truth Ground truth: compare SAM 3 (or MedSAM box from GT bbox) with an expert mask -> Dice, IoU, HD95. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_compare_models Compare: SAM 3 text vs MedSAM (box from SAM 3) vs SAM 3 tracker (same box); pairwise Dice. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_edge_regions Classical Canny-based region proposal (no GPU) for quick boundary exploration. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_research_pipeline Research: batch segmentation with per-subject statistics and a ZIP export bundle. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_segment_interactive Point & Box: SAM 3 tracker (points/box) or MedSAM (box) interactive segmentation. points_text: 'x,y,label; x,y,label' in pixel coordinates (label 1 = foreground, 0 = background). If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_segment_text Segment: SAM 3 text-prompted segmentation with instance table, ROI metrics and exports. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_series_segment Series: segment every ordered slice of the selected series with one prompt; estimate volume. If a user passes a file as an input, use the upload_file_to_gradio tool, if present, to upload the file to the gradio app and create a Gradio File Input. Then use the returned path as the input to the tool
NeuroSAM3_system_diagnostics Run a GPU probe (torch/CUDA info, model devices, raw SAM 3 scores on a bundled FLAIR slice) and return the recent application log.

From Hugging Face Spaces. Report a problem