> ## Documentation Index > Fetch the complete documentation index at: https://docs.aireserve.com/llms.txt > Use this file to discover all available pages before exploring further. # Quote a run > Prices a fine-tuning request from the validated token count and your choices (base, method, serving, ceiling). Answers a signed `quoteId` to start from, or `409 no_placement` with what would work. Requires compute `read` access. ## OpenAPI ````yaml /api-reference/openapi.json post /v1/training/jobs/quote openapi: 3.0.3 info: title: AI Reserve Gateway API description: >- The AI Reserve API surface — OpenAI-compatible chat completions, Anthropic-compatible messages, Bedrock Converse, images, video, files, text-to-speech, realtime voice sessions, provider-hidden GPU infrastructure, and fine-tuning — all behind one endpoint and one API key. version: 1.0.0 servers: - url: https://api.aireserve.com security: - bearerAuth: [] tags: - name: Customers description: >- Per-end-customer usage, cost, and billable amounts for organizations that tag requests with their own customers. - name: GPU Infrastructure description: >- Provision and manage provider-hidden GPU instances. The recommended Instances API accepts workload requirements and an hourly price ceiling; AI Reserve handles provider placement internally. Every mutation is asynchronous and returns an operation to poll. - name: Customer keys description: >- Mint, list, and revoke API keys pinned to your own end customers (and their organizations), so spend splits per customer without per-request tagging. - name: AWS CLI login description: >- Browser-authorization endpoints consumed by `aws login --profile …` (AWS CLI v2.32.0+). All three routes are unauthenticated — the caller has no credential until `/v1/token` succeeds. - name: Fine-tuning description: >- Fine-tune open models on your own data. Data lives in an S3 bucket inside your organization's AWS hub account; you name a base, LoRA or full, a price ceiling and how the result must be servable, and AI Reserve places the run on a backend internally. The result is a model in your bucket you can deploy or download. paths: /v1/training/jobs/quote: post: tags: - Fine-tuning summary: Quote a run description: >- Prices a fine-tuning request from the validated token count and your choices (base, method, serving, ceiling). Answers a signed `quoteId` to start from, or `409 no_placement` with what would work. Requires compute `read` access. operationId: quoteTrainingJob parameters: [] requestBody: required: true content: application/json: schema: type: object properties: name: type: string minLength: 1 maxLength: 120 datasetId: type: string format: uuid pattern: >- ^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$ baseModel: type: string minLength: 1 maxLength: 200 method: type: string enum: - lora - full hyperparameters: default: epochs: 3 learningRate: 0.0001 batchSize: 16 maxSeqLen: 4096 loraRank: 16 type: object properties: epochs: default: 3 type: integer minimum: 1 maximum: 50 learningRate: default: 0.0001 type: number minimum: 0 exclusiveMinimum: true maximum: 1 batchSize: default: 16 type: integer minimum: 1 maximum: 256 maxSeqLen: default: 4096 type: integer minimum: 256 maximum: 131072 loraRank: type: integer minimum: 1 maximum: 128 additionalProperties: false serving: type: string enum: - serverless - dedicated - download maxCostUsd: type: number minimum: 0 maximum: 100000 providerParameters: type: object additionalProperties: anyOf: - type: string maxLength: 512 - type: number - type: boolean required: - name - datasetId - baseModel - method - serving - maxCostUsd additionalProperties: false responses: '200': description: OK content: application/json: schema: type: object properties: quoteId: type: string description: >- Signed quote; pass it to `POST /v1/training/jobs` within 120 seconds. expiresAt: type: string baseModel: type: string method: type: string enum: - lora - full serving: type: string enum: - serverless - dedicated - download datasetId: type: string format: uuid pattern: >- ^([0-9a-fA-F]{8}-[0-9a-fA-F]{4}-[1-8][0-9a-fA-F]{3}-[89abAB][0-9a-fA-F]{3}-[0-9a-fA-F]{12}|00000000-0000-0000-0000-000000000000|ffffffff-ffff-ffff-ffff-ffffffffffff)$ estimatedTokens: type: number estimatedCostUsd: type: string estimatedMinutes: type: number servingPrice: anyOf: - type: object properties: kind: type: string enum: - serverless usdPerMillionInputTokens: type: number usdPerMillionOutputTokens: type: number required: - kind - usdPerMillionInputTokens - usdPerMillionOutputTokens additionalProperties: false - type: object properties: kind: type: string enum: - dedicated usdPerHour: type: number required: - kind - usdPerHour additionalProperties: false - type: object properties: kind: type: string enum: - download required: - kind additionalProperties: false hyperparameters: type: object properties: epochs: default: 3 type: integer minimum: 1 maximum: 50 learningRate: default: 0.0001 type: number minimum: 0 exclusiveMinimum: true maximum: 1 batchSize: default: 16 type: integer minimum: 1 maximum: 256 maxSeqLen: default: 4096 type: integer minimum: 256 maximum: 131072 loraRank: type: integer minimum: 1 maximum: 128 required: - epochs - learningRate - batchSize - maxSeqLen additionalProperties: false description: >- As they will be submitted: the request's values clamped into the placed backend's range. adjustments: type: array items: type: object properties: field: type: string enum: - batchSize - loraRank - maxSeqLen from: type: number to: type: number reason: type: string required: - field - from - to - reason additionalProperties: false description: >- Every hyperparameter placement changed to fit the backend's published range, with the reason. required: - quoteId - expiresAt - baseModel - method - serving - datasetId - estimatedTokens - estimatedCostUsd - estimatedMinutes - servingPrice - hyperparameters - adjustments additionalProperties: false '400': description: Invalid request or schema violation. content: application/json: schema: $ref: '#/components/schemas/TrainingError' '401': description: Missing or invalid API key. content: application/json: schema: $ref: '#/components/schemas/TrainingError' '402': description: >- Wallet not enabled or cannot cover the estimate (`insufficient_funds`). content: application/json: schema: $ref: '#/components/schemas/TrainingError' '403': description: Key lacks the required compute access (`read` or `manage`). content: application/json: schema: $ref: '#/components/schemas/TrainingError' '404': description: Resource not found or owned by another organization. content: application/json: schema: $ref: '#/components/schemas/TrainingError' '409': description: >- `no_placement` with `details` (see `TrainingNoPlacementDetails`), `dataset_not_ready`, `quote_expired`, or `aws_account_required`. content: application/json: schema: $ref: '#/components/schemas/TrainingError' '503': description: >- `storage_unconfigured` or `backend_unconfigured`: the environment is missing a credential; retry after it is added. content: application/json: schema: $ref: '#/components/schemas/TrainingError' security: - bearerAuth: [] - apiKeyAuth: [] components: schemas: TrainingError: type: object properties: error: type: object properties: message: type: string type: type: string enum: - invalid_request_error param: type: string nullable: true enum: - null code: type: string details: type: object additionalProperties: {} required: - message - type - param - code additionalProperties: false required: - error additionalProperties: false securitySchemes: bearerAuth: type: http scheme: bearer description: >- Your AI Reserve API key (aireserve_api_…) from the API Keys page in the portal (https://portal.aireserve.com/keys). apiKeyAuth: type: apiKey in: header name: x-api-key description: The same API key sent as an `x-api-key` header. ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.