API Schemas for Authentication, AI Models, and Model Creation

12th chunk of `docs/ragithub/openapi.yaml`

06c10daf269c8c44f7616b3b5177729e77844ba7432e3ccf0000000100000e7a

 multi_turn_schema:
          type: object
          description: Ragit re-writes queries when there are multiple turns. It's the actual query that's used for RAG.
          nullable: true
          properties:
            is_query:
              type: boolean
            in_context:
              type: boolean
            query:
              type: string
        created_at:
          type: string
          format: date-time
    ApiKey:
      type: object
      properties:
        api_key_preview:
          type: string
          description: first 9 characters of the api key
        name:
          type: string
        expire:
          type: string
          format: date-time
    ApiKeyCreation:
      type: object
      properties:
        name:
          type: string
        expire_after:
          type: integer
          description: days
        password:
          type: string
          description: user password in plain text
    AiModel:
      type: object
      properties:
        id:
          type: string
          description: "It's a unique identifier of the model. It's a hash value of `name` + `api_name` + `api_provider` + `api_url`."
        name:
          type: string
          description: It's a name for humans.
        api_name:
          type: string
          description: It's a name for apis.
        api_provider:
          type: string
          description: openai | anthropic | cohere | google
        api_url:
          type: string
          nullable: true
          description: openai-compatible apis need this field. If it's not set, it uses the default url of each api provider.
        can_read_images:
          type: boolean
        input_price:
          type: number
          format: double
          description: dollars per 1 million input tokens
        output_price:
          type: number
          format: double
          description: dollars per 1 million output tokens
        explanation:
          type: string
          nullable: true
        api_env_var:
          type: string
          nullable: true
          description: Ragit uses this environment variable for api key.
          example: OPENAI_API_KEY
        tags:
          type: array
          description: It helps users find models more easily in ragithub.
          items:
            type: string
        created_at:
          type: string
          format: date-time
        updated_at:
          type: string
          format: date-time
    AiModelCreation:
      type: object
      properties:
        name:
          type: string
          description: It's a name for humans.
        api_name:
          type: string
          description: It's a name for apis.
        api_provider:
          type: string
          description: openai | anthropic | cohere | google
        api_url:
          type: string
          nullable: true
          description: openai-compatible apis need this field. If it's not set, it uses the default url of each api provider.
        can_read_images:
          type: boolean
        input_price:
          type: number
          format: double
          description: dollars per 1 million input tokens
        output_price:
          type: number
          format: double
          description: dollars per 1 million output tokens
        explanation:
          type: string
          nullable: true
        api_env_var:
          type: string
          nullable: true
          description: Ragit uses this environment variable for api key.
          example: OPENAI_API_KEY
        tags:
          type: array
          description: It helps users find models more easily in ragithub.
          items:
            type: string

Title: API Schemas for Authentication, AI Models, and Model Creation

Summary

This section defines API schemas for API keys, AI models, and AI model creation, including properties for authentication, model identification, pricing, and environmental variables, as well as details for creating new AI models with specific parameters and tags.