> ## Documentation Index > Fetch the complete documentation index at: https://docs.together.ai/llms.txt > Use this file to discover all available pages before exploring further. # Estimate price > Estimate the price of a fine-tuning job. ## OpenAPI ````yaml POST /fine-tunes/estimate-price openapi: 3.1.0 info: title: Together APIs description: The Together REST API. See https://docs.together.ai for more details. version: 2.0.0 termsOfService: https://www.together.ai/terms-of-service contact: name: Together Support url: https://www.together.ai/contact license: name: MIT url: https://github.com/togethercomputer/openapi/blob/main/LICENSE servers: - url: https://api.together.ai/v1 description: Default environment for APIs - url: https://api-inference.together.ai/v2 description: Optimized environment for inference security: - bearerAuth: [] paths: /fine-tunes/estimate-price: post: tags: - Fine-tuning summary: Estimate price description: Estimate the price of a fine-tuning job. requestBody: required: true content: application/json: schema: type: object required: - training_file properties: training_file: type: string description: File-ID of a training file uploaded to the Together API validation_file: type: string description: File-ID of a validation file uploaded to the Together API model: type: string description: Name of the base model to run fine-tune job on n_epochs: type: integer default: 1 description: >- Number of complete passes through the training dataset (higher values may improve results but increase cost and risk of overfitting) n_evals: type: integer default: 0 description: >- Number of evaluations to be run on a given validation set during training training_method: type: object oneOf: - $ref: '#/components/schemas/TrainingMethodSFT' - $ref: '#/components/schemas/TrainingMethodDPO' description: >- The training method to use. 'sft' for Supervised Fine-Tuning or 'dpo' for Direct Preference Optimization. training_type: type: object default: null nullable: true oneOf: - $ref: '#/components/schemas/FullTrainingType' - $ref: '#/components/schemas/LoRATrainingType' description: The training type to use. Defaults to LoRA if not provided. from_checkpoint: type: string description: >- The checkpoint identifier to continue training from a previous fine-tuning job. Format is `{$JOB_ID}` or `{$OUTPUT_MODEL_NAME}` or `{$JOB_ID}:{$STEP}` or `{$OUTPUT_MODEL_NAME}:{$STEP}`. The step value is optional; without it, uses the final checkpoint. responses: '200': description: Price estimated successfully content: application/json: schema: oneOf: - type: object x-stainless-variantName: AvailableEstimate required: - estimation_available properties: estimation_available: type: boolean enum: - true x-stainless-const: true description: >- Whether price estimation is available for the requested fine-tune job. estimated_total_price: type: number description: The price of the fine-tuning job allowed_to_proceed: type: boolean description: >- Whether you are allowed to proceed with the fine-tuning job. example: true user_limit: type: number description: Your credit limit in dollars. estimated_train_token_count: type: number description: The estimated number of tokens to be trained estimated_eval_token_count: type: number description: The estimated number of tokens for evaluation - type: object x-stainless-variantName: UnavailableEstimate required: - estimation_available - unavailable_reason properties: estimation_available: type: boolean enum: - false x-stainless-const: true description: >- Whether price estimation is available for the requested fine-tune job. unavailable_reason: type: string description: >- Reason price estimation is unavailable for the requested fine-tune job. enum: - multimodal_dataset - train_file_not_validated - eval_file_not_validated - train_file_invalid - eval_file_invalid example: multimodal_dataset '500': description: Internal Server Error content: application/json: schema: $ref: '#/components/schemas/ErrorData' components: schemas: TrainingMethodSFT: type: object properties: method: type: string enum: - sft train_on_inputs: oneOf: - type: boolean - type: string enum: - auto type: boolean default: auto description: >- Whether to mask user messages in conversational data or prompts in instruction data. required: - method - train_on_inputs TrainingMethodDPO: type: object properties: method: type: string enum: - dpo dpo_beta: type: number default: 0.1 rpo_alpha: type: number default: 0 dpo_normalize_logratios_by_length: type: boolean default: false dpo_reference_free: type: boolean default: false simpo_gamma: type: number default: 0 required: - method FullTrainingType: type: object properties: type: type: string enum: - Full required: - type LoRATrainingType: type: object properties: type: type: string enum: - Lora lora_r: type: integer lora_alpha: type: integer lora_dropout: type: number default: 0 lora_trainable_modules: type: string default: all-linear description: >- Comma-separated LoRA target modules. Use `all-linear` for model defaults; MoE expert modules (`w_up`, `w_gate`, `w_down`) are supported on compatible models and cannot be mixed with attention modules. required: - type - lora_r - lora_alpha ErrorData: type: object required: - error properties: error: type: object properties: message: type: string nullable: false type: type: string nullable: false param: type: string nullable: true default: null code: type: string nullable: true default: null required: - type - message securitySchemes: bearerAuth: type: http scheme: bearer x-bearer-format: bearer x-default: default ````