Skip to main content
Glama

create

Generate custom AI models locally by defining parameters in a Modelfile, enabling tailored machine learning solutions for specific applications.

Instructions

Create a model from a Modelfile

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
nameYesName for the model
modelfileYesPath to Modelfile

Implementation Reference

  • The function that executes the 'create' tool by running the 'ollama create' command using the provided model name and Modelfile path.
    private async handleCreate(args: any) { try { const { stdout, stderr } = await execAsync(`ollama create ${args.name} -f ${args.modelfile}`); return { content: [ { type: 'text', text: stdout || stderr, }, ], }; } catch (error) { throw new McpError(ErrorCode.InternalError, `Failed to create model: ${formatError(error)}`); } }
  • Input schema defining the required 'name' and 'modelfile' parameters for the 'create' tool.
    inputSchema: { type: 'object', properties: { name: { type: 'string', description: 'Name for the model', }, modelfile: { type: 'string', description: 'Path to Modelfile', }, }, required: ['name', 'modelfile'], additionalProperties: false, },
  • src/index.ts:76-94 (registration)
    Registration of the 'create' tool in the ListTools response, including name, description, and input schema.
    { name: 'create', description: 'Create a model from a Modelfile', inputSchema: { type: 'object', properties: { name: { type: 'string', description: 'Name for the model', }, modelfile: { type: 'string', description: 'Path to Modelfile', }, }, required: ['name', 'modelfile'], additionalProperties: false, }, },
  • src/index.ts:258-259 (registration)
    Dispatch in CallToolRequestHandler that routes 'create' tool calls to the handleCreate method.
    case 'create': return await this.handleCreate(request.params.arguments);

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/NightTrek/Ollama-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server