openapi: 3.1.0 info: title: Flux Training API description: API endpoints for training, monitoring, and managing Flux models (AI image models) version: '1.0.0' servers: - url: https://training-suite.simplismart.ai description: Training Suite Production Server tags: - name: Flux Training description: Flux model training endpoints paths: /api/flux/train/: post: tags: - Flux Training summary: Start a new Flux training job description: Submit a new Flux training job with the specified configuration, training data, and metadata. operationId: startFluxTrainingJob parameters: - in: header name: Authorization schema: type: string required: true description: Bearer token for authentication and authorization. example: Bearer requestBody: required: true content: multipart/form-data: schema: type: object properties: trigger_word: type: string description: The trigger word to be associated with generated images. example: "portrait of a person" rank: type: string description: The rank parameter for training. example: "128" steps: type: string description: Number of training steps. example: "1000" lr: type: string description: Learning rate for training. example: "0.0001" optimizers: type: string description: Optimizer to use for training. example: "Adam" high_resolution_mode: type: string description: Whether to enable high resolution mode. example: "false" org: type: string description: Organization ID associated with the training job. example: "0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d" dataset_file: type: string format: binary description: Dataset file containing training images. name: type: string description: Name assigned to the Flux training job. example: "my_first_flux_job" required: - trigger_word - dataset_file - org - name responses: '200': description: Flux training job submitted successfully. content: application/json: schema: $ref: '#/components/schemas/FluxTrainingJobResponse' /api/flux/training_job/list/: get: tags: - Flux Training summary: Retrieve a list of all Flux training jobs description: Fetches a list of all Flux training jobs associated with the specified organization. Each job includes details such as its ID, name, status, and other metadata. operationId: listFluxTrainingJobs parameters: - in: query name: org_id schema: type: string required: true description: The unique identifier of the organization for which the training jobs are being retrieved. example: 0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d - in: header name: Authorization schema: type: string required: true description: The authorization token used to authenticate the request. It must be provided in the "Bearer {token}" format. example: Bearer responses: '200': description: Successful response containing a list of Flux training jobs. content: application/json: schema: $ref: '#/components/schemas/ListFluxTrainingJobsResponse' '400': description: Invalid request, such as missing or incorrect parameters. '401': description: Unauthorized access due to invalid or missing authorization token. '500': description: Internal server error while retrieving the training jobs. /api/flux/training_job/get/: get: tags: - Flux Training summary: Retrieve details of a specific Flux training job description: Fetch metadata and configuration details of a Flux training job identified by the provided lora_id and org_id. operationId: getFluxTrainingJob parameters: - in: query name: org_id schema: type: string required: true description: Organization ID to which the training job belongs. example: 0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d - in: query name: lora_id schema: type: string required: true description: Unique identifier for the Flux training job. example: 3fe92c25-e5fd-4281-93df-5aa886064dbd - in: header name: Authorization schema: type: string required: true description: Bearer token for authentication and authorization. example: Bearer responses: '200': description: The requested Flux training job details were retrieved successfully. content: application/json: schema: $ref: '#/components/schemas/GetFluxTrainingJobResponse' /api/flux/training_job/get/mlflow-metrics/: get: tags: - Flux Training summary: Retrieve Flux training metrics description: This endpoint allows you to fetch performance metrics for a specific Flux training job. Metrics include training loss and other relevant statistics logged during the training process. operationId: getFluxTrainingMetrics parameters: - in: query name: org_id schema: type: string required: true description: The unique identifier of the organization for which the metrics are being requested. example: 0bf00b43-430a-4ca3-a8b3-b13cc8dc6d4d - in: query name: request_id schema: type: string required: true description: A unique identifier corresponding to the training job whose metrics are being fetched. example: c3c840b8-aee0-4b76-9da8-0ac4e0c0df8e - in: header name: Authorization schema: type: string required: true description: The authorization token used to authenticate the request. It should be provided in the "Bearer {token}" format. example: Bearer responses: '200': description: Successful response containing the requested metrics. content: application/json: schema: $ref: '#/components/schemas/FluxMlflowMetricsResponse' '400': description: Invalid request, such as missing or incorrect parameters. '401': description: Unauthorized access due to invalid or missing authorization token. '404': description: Training job or metrics not found for the provided request_id and org_id. '500': description: Internal server error while retrieving the metrics. components: schemas: ListFluxTrainingJobsResponse: type: object properties: jobs: type: array items: $ref: '#/components/schemas/FluxTrainingJobSummary' FluxTrainingJobSummary: type: object properties: lora_id: type: string description: Unique identifier for the Flux training job name: type: string description: Name of the Flux training experiment status: type: string description: Current status of the training job enum: [QUEUED, RUNNING, COMPLETED, FAILED, CANCELED] created_at: type: string format: date-time description: Timestamp when the job was created trigger_word: type: string description: The trigger word associated with generated images steps: type: string description: Number of training steps configured FluxTrainingJobResponse: type: object properties: lora_id: type: string description: Unique identifier for the Flux training job status: type: string description: Initial status of the training job message: type: string description: Additional information about the job submission GetFluxTrainingJobResponse: type: object properties: lora_id: type: string description: Unique identifier for the Flux training job name: type: string description: Name of the Flux training experiment status: type: string description: Current status of the training job created_at: type: string format: date-time description: Timestamp when the job was created config: type: object description: The complete training configuration used for this job progress: type: number description: Progress percentage of the training job (0-100) trigger_word: type: string description: The trigger word associated with this model steps: type: string description: Number of training steps configured learning_rate: type: string description: Learning rate used for training output_model: type: string description: Path to the trained model output (when completed) FluxMlflowMetricsResponse: type: object properties: metrics: type: array items: $ref: '#/components/schemas/FluxMetricEntry' FluxMetricEntry: type: object properties: key: type: string description: Name of the metric value: type: number description: Value of the metric timestamp: type: string format: date-time description: Timestamp when the metric was recorded step: type: integer description: Training step at which the metric was recorded securitySchemes: BearerAuth: type: http scheme: bearer bearerFormat: JWT description: JWT token for authentication security: - BearerAuth: []