charmcraft.yaml

# This file configures Charmcraft.
# See https://juju.is/docs/sdk/charmcraft-config for guidance.

name: ollama
type: charm
title: Charmed Ollama

summary: Deploys and manages Ollama, an inference engine for large language models.

description: |
  Get up and running with large language models.

  Ollama is an inference engine for large language models.
  This charm is suitable for developers and organizations looking to harness the power of large language models in their applications without the complexity of manual setup and configuration.

  Ollama exposes an HTTP API with the same format as OpenAI, enabling rapid migration of your application from closed-source models like OpenAI's to open-source models like Llama and Mistral.

bases:
  - build-on:
    - name: ubuntu
      channel: "22.04"
    run-on:
    - name: ubuntu
      channel: "22.04"

config:
  options:
    port:
      description: The port on which Ollama will expose its API.
      default: 11434
      type: int

actions:
  generate:
    description: |
      Completes a given prompt using the large language model you specify.
      Notice: you need to pull the model you want to use prior to running the "generate" action.
    params:
      model:
        type: string
        description: "The name of the model to use"
      prompt:
        type: string
        description: "The prompt you want to give to the LLM"
    required:
    - prompt
    additionalProperties: false
  pull:
    description: "Downloads and serves the model you specify. Find available models here: https://ollama.com/library"
    params:
      model:
        type: string
        description: "The name of the model to pull"
    required:
    - model
    additionalProperties: false