mcpField
J
jevjam

jevjam

Local
github.comAI & Agents
RepositoryVisit website

About jevjam

jevjam is a self-hosted MCP server and HTTP API for typed decisions using Laya, Julia-1, and clef-flash on an NVIDIA GPU. It serves agents and services needing choices, scores, or yes/no judgments from text, JSON, images, or video.

Key features

  1. 01Choice, score, and yes/no predictions
  2. 02MCP tools and HTTP API
  3. 03Laya, Julia-1, and clef-flash models
  4. 04Model loading and idle GPU memory release
  5. 05Optional bearer-key authentication

Use cases

  1. 01Routing support tickets by department
  2. 02Flagging potentially abusive content
  3. 03Checking risk before allowing a tool call
  4. 04Selecting a model for a prompt

MCP support

Exposes four MCP tools: `jevjam_predict`, `jevjam_preset`, `jevjam_route`, and `jevjam_status`. MCP and HTTP share a queue and resident model.

Configuration

Local server config is the same for every client. Copy it into your client’s MCP config file.

mcp.json · Universal config
{
  "mcpServers": {
    "beremaran-jevjam": {
      "command": "docker",
      "args": [
        "run",
        "-i",
        "--rm",
        "-e",
        "JEVJAM_API_KEY",
        "-e",
        "JEVJAM_IDLE_TIMEOUT",
        "-e",
        "HF_TOKEN",
        "ghcr.io/beremaran/jevjam:0.3.3"
      ],
      "env": {
        "JEVJAM_API_KEY": "<JEVJAM_API_KEY>",
        "JEVJAM_IDLE_TIMEOUT": "<JEVJAM_IDLE_TIMEOUT>",
        "HF_TOKEN": "<HF_TOKEN>"
      }
    }
  }
}
Run commanddocker run -i --rm -e JEVJAM_API_KEY -e JEVJAM_IDLE_TIMEOUT -e HF_TOKEN ghcr.io/beremaran/jevjam:0.3.3
Environment variables to set
JEVJAM_API_KEYSecretWhen set, /mcp and /v1/systemone need 'Authorization: Bearer <key>'
JEVJAM_IDLE_TIMEOUTSeconds without inference before the resident checkpoint is freed; 0 keeps it
HF_TOKENSecretOptional Hugging Face token for higher download rate limits

Local servers need a runtime on your machine (e.g. Node.js 18+). Config file locations differ by client; see your client’s docs.

Packages
ocighcr.io/beremaran/jevjam:0.3.3

Tools

No tool list yet

Local servers expose tools only once running on your machine, and the maintainer has not provided a tool list.

Details

The maintainer has not added details yet. Check the README in the repository.

Comments

Markdown supported

New to MCP?

From what the protocol is to installing your first server, these answers cover most first questions.

An open protocol that defines how AI applications connect to external tools and data in a uniform way. Clients talk to the model; MCP servers provide tools, resources and prompt templates. They exchange standard messages, so one server can be reused by many clients.