> ## Documentation Index > Fetch the complete documentation index at: https://docs.platform.toloka.ai/docs/llms.txt > Use this file to discover all available pages before exploring further. # Resume a training run > Retry a run blocked on funds or payment after adding funds. Returns its updated state. Returns 409 if the run is not resumable or its payment window has expired. ## OpenAPI ````yaml /api-reference/openapi.json post /training-runs/{trainingRunId}/resume openapi: 3.1.0 info: description: >- Manage projects, datasets, labeling pipelines, and model training. **Beta:** endpoints, request and response shapes, and error contracts can change without notice or a backward-compatibility promise. title: Toloka Platform API version: 2.0.0 servers: - url: https://platform.toloka.ai/api/v2-beta security: - apiKeyAuth: [] - scopedTokenAuth: [] tags: - description: The authenticated API key owner. name: Account - description: A project is the container for pipelines, datasets, and files. name: Projects - description: Rows and typed fields. Field schemas are immutable once created. name: Datasets - description: Append rows to a dataset from a file. name: Dataset uploads - description: Read a dataset out as a file. name: Dataset exports - description: A pipeline is a graph of nodes that processes each dataset item. name: Pipelines - description: The nodes, connections, fields, and version of a pipeline graph. name: Pipeline graph - description: Execute a pipeline over a dataset and inspect each item. name: Runs - description: Per-node checks that a pipeline must pass before deployment. name: Checks - description: Named credentials available to pipeline code. name: Secrets - description: Project files available to pipeline code and labeling interfaces. name: Files - description: >- Global, read-only catalog of the models a synthetic generation node can use, with their budget bounds. name: Generation models - description: >- Global, read-only catalog of the expert audience: the values a node `audience` filter can use. name: Audience - description: >- Project lists of experts a node can take its workforce from with `audienceMode: service`. Available only to teams with the audience service enabled. name: Expert lists - description: >- Experts banned in a project: they cannot be invited and are removed from the work they are in. Available only to teams with expert bans enabled. name: Expert ban list - description: Global, read-only base model catalog. name: Base models - description: Project-owned mutable model-training drafts. name: Trainings - description: Project-owned Models and immutable Model Versions. name: Models - name: Batch inference paths: /training-runs/{trainingRunId}/resume: post: tags: - Trainings summary: Resume a training run description: >- Retry a run blocked on funds or payment after adding funds. Returns its updated state. Returns 409 if the run is not resumable or its payment window has expired. operationId: resumeTrainingRun parameters: - in: path name: trainingRunId required: true schema: $ref: '#/components/schemas/TrainingRunId' requestBody: content: application/json: example: {} schema: additionalProperties: false example: {} properties: {} type: object required: true responses: '200': content: application/json: schema: $ref: '#/components/schemas/TrainingRun' description: Resumed Training Run headers: RateLimit-Limit: $ref: '#/components/headers/RateLimit-Limit' RateLimit-Policy: $ref: '#/components/headers/RateLimit-Policy' RateLimit-Remaining: $ref: '#/components/headers/RateLimit-Remaining' RateLimit-Reset: $ref: '#/components/headers/RateLimit-Reset' X-Request-Id: $ref: '#/components/headers/X-Request-Id' '409': description: Training Run is not resumable from its current state headers: RateLimit-Limit: $ref: '#/components/headers/RateLimit-Limit' RateLimit-Policy: $ref: '#/components/headers/RateLimit-Policy' RateLimit-Remaining: $ref: '#/components/headers/RateLimit-Remaining' RateLimit-Reset: $ref: '#/components/headers/RateLimit-Reset' X-Request-Id: $ref: '#/components/headers/X-Request-Id' '429': description: Rate limit exceeded headers: RateLimit-Limit: $ref: '#/components/headers/RateLimit-Limit' RateLimit-Policy: $ref: '#/components/headers/RateLimit-Policy' RateLimit-Remaining: $ref: '#/components/headers/RateLimit-Remaining' RateLimit-Reset: $ref: '#/components/headers/RateLimit-Reset' X-Request-Id: $ref: '#/components/headers/X-Request-Id' x-codeSamples: - label: TypeScript SDK lang: typescript source: | import { createClient } from "@toloka-platform/sdk"; const client = createClient({ baseUrl: "https://platform.toloka.ai", apiKey: process.env.TOLOKA_API_KEY!, }); const run = await client.trainings.resumeRun({ trainingRunId: "trn.019a0000-0000-7000-8000-000000000001" }); console.log(run); components: schemas: TrainingRunId: description: ID of the training run, with prefix `trn.`. pattern: >- ^trn\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string TrainingRun: description: The configuration, inputs, and current state of one training run. properties: billingStatus: description: >- Billing state: `pending` before reservation, `reserved` while funds are held, `release_pending` or `released` for returned funds, `settled` after charging, `payment_required` for a shortfall, `resolved` after staff review, or `reconciliation_required` when charges need review. enum: - pending - reserved - release_pending - released - settled - payment_required - resolved - reconciliation_required type: string blockedReason: anyOf: - enum: - funds - payment - billing_review type: string - type: 'null' description: >- Reason the run is blocked: `funds` to reserve the budget, `payment` to settle usage, or `billing_review` for staff review; otherwise null. chargedAmount: anyOf: - pattern: ^(0|[1-9][0-9]*)(\.[0-9]+)?$ type: string - type: 'null' description: >- Settled charge in US dollars as a decimal string, or null before a charge is recorded. configuration: $ref: '#/components/schemas/TrainingConfiguration' createdAt: description: Time the training run was created, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string error: anyOf: - additionalProperties: false properties: code: anyOf: - enum: - data_export_failed - input_report_mismatch - billing_rejected - provider_failed - provider_timeout - invalid_provider_output - provider_output_changed - prompt_changed - payment_window_expired type: string - const: canceled type: string description: Machine-readable reason the run failed or was canceled. message: description: Explanation of the failure or cancellation. maxLength: 500 type: string retryable: description: Whether this failure is marked as retryable. type: boolean required: - code - message - retryable type: object - type: 'null' description: Failure or cancellation details, or null when no error is recorded. estimateFingerprint: description: >- SHA-256 fingerprint of the accepted estimate, as 64 lowercase hexadecimal characters. pattern: ^[0-9a-f]{64}$ type: string estimateInputSetId: $ref: '#/components/schemas/TrainingCostEstimateInputSetId' id: $ref: '#/components/schemas/TrainingRunId' inputCutoff: description: Cutoff used to freeze the input rows, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string inputs: anyOf: - additionalProperties: false properties: training: additionalProperties: false description: Prepared files and validation report for the training input. properties: byteSize: description: Size of the prepared input file in bytes. maximum: 9007199254740991 minimum: 0 type: integer contract: description: Versioned format contract for the training examples. enum: - openai-chat-v1 type: string datasetExportId: description: >- Export ID with prefix `dex.` that prepared these examples. pattern: >- ^dex\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetFileId: description: Prepared dataset file ID with prefix `dfl.`. pattern: >- ^dfl\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetId: description: Dataset ID with prefix `ds.` containing these examples. pattern: >- ^ds\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string format: const: training-jsonl description: >- File format; `training-jsonl` stores one training example per JSON line. type: string itemFilter: $ref: '#/components/schemas/TrainingInputItemFilter' report: $ref: '#/components/schemas/TrainingInputReport' description: >- Validation results and token counts for the prepared examples. sha256: description: >- SHA-256 checksum of the input file, as 64 lowercase hexadecimal characters. pattern: ^[0-9a-f]{64}$ type: string required: - datasetId - datasetExportId - datasetFileId - itemFilter - format - contract - byteSize - sha256 - report type: object validation: anyOf: - additionalProperties: false properties: byteSize: description: Size of the prepared input file in bytes. maximum: 9007199254740991 minimum: 0 type: integer contract: description: Versioned format contract for the training examples. enum: - openai-chat-v1 type: string datasetExportId: description: >- Export ID with prefix `dex.` that prepared these examples. pattern: >- ^dex\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetFileId: description: Prepared dataset file ID with prefix `dfl.`. pattern: >- ^dfl\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetId: description: >- Dataset ID with prefix `ds.` containing these examples. pattern: >- ^ds\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string format: const: training-jsonl description: >- File format; `training-jsonl` stores one training example per JSON line. type: string itemFilter: $ref: '#/components/schemas/TrainingInputItemFilter' report: $ref: '#/components/schemas/TrainingInputReport' description: >- Validation results and token counts for the prepared examples. sha256: description: >- SHA-256 checksum of the input file, as 64 lowercase hexadecimal characters. pattern: ^[0-9a-f]{64}$ type: string required: - datasetId - datasetExportId - datasetFileId - itemFilter - format - contract - byteSize - sha256 - report type: object - type: 'null' description: >- Prepared validation input, or null when no validation input was selected. required: - training - validation type: object - type: 'null' description: >- Prepared training and validation inputs; null when unavailable or omitted from create and list responses. maxCost: description: >- Maximum accepted charge for this run, in US dollars as a decimal string. type: string message: description: >- Current status or billing explanation, or null when none is available. type: - string - 'null' outputModelVersionId: $ref: '#/components/schemas/ModelVersionId' outputModelVersionName: description: Name reserved for the model version this run produces. maxLength: 120 minLength: 1 type: string outputModelVersionRevision: anyOf: - exclusiveMinimum: 0 maximum: 9007199254740991 type: integer - type: 'null' description: Published model version revision, or null before publication. progress: anyOf: - additionalProperties: false properties: message: anyOf: - maxLength: 500 type: string - type: 'null' description: Progress explanation, or null when none is available. phase: description: Current phase of training. maxLength: 120 minLength: 1 type: string step: anyOf: - maximum: 9007199254740991 minimum: 0 type: integer - type: 'null' description: >- Current training step, or null when step progress is unavailable. total: anyOf: - exclusiveMinimum: 0 maximum: 9007199254740991 type: integer - type: 'null' description: Total number of training steps, or null when unknown. required: - phase - step - total - message type: object - type: 'null' description: >- Latest training progress while running or stopping; null in other states or when progress is unavailable. quote: description: Cost estimate accepted for this run. properties: budget: $ref: '#/components/schemas/TrainingResourceBudget' calculatedAt: description: Time the estimate was calculated, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string capabilityVersion: description: Version of the model capability rules used for this estimate. minLength: 1 type: string currency: const: USD description: Currency used for the estimate; always `USD`. type: string cutoff: description: >- Dataset cutoff used to select the inputs, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string inputSetId: $ref: '#/components/schemas/TrainingCostEstimateInputSetId' inputs: $ref: '#/components/schemas/TrainingCostEstimateInputs' maximumCharge: description: Maximum charge in US dollars as a decimal string. type: string pricingVersion: description: Version of the pricing rules used for this estimate. minLength: 1 type: string required: - inputSetId - cutoff - inputs - capabilityVersion - pricingVersion - calculatedAt - budget - maximumCharge - currency type: object status: description: >- Run state: `queued`, `running`, `stopping`, `blocked`, `completed`, `failed`, or `canceled`; `blocked` requires resolving the reported reason. enum: - queued - running - stopping - blocked - completed - failed - canceled type: string trainingId: $ref: '#/components/schemas/TrainingId' updatedAt: description: Time the training run was last updated, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string required: - id - trainingId - configuration - estimateInputSetId - estimateFingerprint - outputModelVersionId - outputModelVersionName - maxCost - status - outputModelVersionRevision - progress - blockedReason - error - billingStatus - chargedAmount - message - quote - inputCutoff - inputs - createdAt - updatedAt type: object TrainingConfiguration: description: A normalized method-specific Training draft. oneOf: - $ref: '#/components/schemas/SftTrainingConfiguration' - $ref: '#/components/schemas/GistingTrainingConfiguration' TrainingCostEstimateInputSetId: description: The immutable input set used to calculate a Training Cost Estimate. pattern: >- ^tei\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string TrainingInputItemFilter: additionalProperties: false description: Filter used to select this frozen training or validation input. properties: createdBefore: description: Include rows created before this ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string excludeParts: $ref: '#/components/schemas/DatasetSplitPartExclusions' fieldSources: $ref: '#/components/schemas/DatasetFieldSourceConditions' fieldValues: $ref: '#/components/schemas/DatasetFieldValueConditions' missingRevisionFields: description: >- Field IDs with prefix `df.` that must have no revision on a selected row; defaults to an empty array. items: pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string type: array parts: $ref: '#/components/schemas/DatasetSplitPartInclusion' slice: $ref: '#/components/schemas/DatasetSliceSelection' required: - createdBefore type: object TrainingInputReport: additionalProperties: false properties: errors: description: Sample of validation errors, up to 50 entries. items: additionalProperties: false properties: itemIndex: description: Index of the dataset item with a validation error. maximum: 9007199254740991 minimum: 0 type: integer reason: description: Reason the example failed validation. type: string required: - itemIndex - reason type: object maxItems: 50 type: array invalidCount: description: Number of examples rejected during validation. maximum: 9007199254740991 minimum: 0 type: integer tokens: additionalProperties: false description: Token-count statistics for the valid examples. properties: estimatorVersion: anyOf: - enum: - utf16-code-units-div-4-v1 - script-weighted-v1 type: string - type: 'null' default: null description: Version of the token estimator used, or null when unavailable. max: description: Largest token count among valid examples. maximum: 9007199254740991 minimum: 0 type: integer p50: description: Median number of tokens per valid example. maximum: 9007199254740991 minimum: 0 type: integer p95: description: 95th percentile of tokens per valid example. maximum: 9007199254740991 minimum: 0 type: integer total: anyOf: - maximum: 9007199254740991 minimum: 0 type: integer - type: 'null' default: null description: Total tokens across valid examples, or null when unavailable. required: - p50 - p95 - max type: object validCount: description: Number of valid training examples in the selection. maximum: 9007199254740991 minimum: 0 type: integer required: - validCount - invalidCount - errors - tokens type: object ModelVersionId: description: Model version ID with prefix `mdv.`. pattern: >- ^mdv\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string TrainingResourceBudget: additionalProperties: false description: Estimated and maximum training resource usage. properties: estimatedGpuSeconds: description: Estimated total GPU time, in seconds summed across all GPUs. maximum: 9007199254740991 minimum: 0 type: integer estimatedWallSeconds: description: Estimated elapsed training time, in seconds. maximum: 9007199254740991 minimum: 0 type: integer gpuCount: description: Number of GPUs used for the estimate. exclusiveMinimum: 0 maximum: 9007199254740991 type: integer maximumGpuSeconds: description: >- Maximum GPU time used to bound the charge, in seconds summed across all GPUs. maximum: 9007199254740991 minimum: 0 type: integer maximumWallSeconds: description: Conservative maximum elapsed training time, in seconds. maximum: 9007199254740991 minimum: 0 type: integer methodWorkUnits: description: >- Token-based work estimate, including epochs and the method-specific training and validation workload. maximum: 9007199254740991 minimum: 0 type: integer required: - methodWorkUnits - estimatedGpuSeconds - estimatedWallSeconds - maximumWallSeconds - maximumGpuSeconds - gpuCount type: object TrainingCostEstimateInputs: additionalProperties: false description: Prepared inputs and their validation reports. properties: training: additionalProperties: false description: Frozen training selection and its validation result. properties: datasetExportId: description: Export ID with prefix `dex.` for the frozen selection. pattern: >- ^dex\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetId: description: Dataset ID with prefix `ds.` containing these examples. pattern: >- ^ds\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string failure: $ref: '#/components/schemas/TrainingInputExportFailure' heldOutRows: description: >- Number of rows excluded by split roles; absent when that count has not been calculated. maximum: 9007199254740991 minimum: 0 type: integer report: anyOf: - $ref: '#/components/schemas/TrainingInputReport' - type: 'null' description: >- Validation report for these examples, or null when the export failed or produced no report. required: - datasetId - datasetExportId - report - failure type: object validation: anyOf: - additionalProperties: false properties: datasetExportId: description: Export ID with prefix `dex.` for the frozen selection. pattern: >- ^dex\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string datasetId: description: Dataset ID with prefix `ds.` containing these examples. pattern: >- ^ds\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string failure: $ref: '#/components/schemas/TrainingInputExportFailure' heldOutRows: description: >- Number of rows excluded by split roles; absent when that count has not been calculated. maximum: 9007199254740991 minimum: 0 type: integer report: anyOf: - $ref: '#/components/schemas/TrainingInputReport' - type: 'null' description: >- Validation report for these examples, or null when the export failed or produced no report. required: - datasetId - datasetExportId - report - failure type: object - type: 'null' description: >- Frozen validation selection and its validation result, or null when no validation input is selected. required: - training - validation type: object TrainingId: description: Training draft ID with prefix `trg.`. pattern: >- ^trg\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string SftTrainingConfiguration: additionalProperties: false description: A normalized supervised fine-tuning draft using LoRA. properties: batchSize: description: >- Sequences per optimizer step across all GPUs: 8, 16, 32, or 64. The default is 8. enum: - 8 - 16 - 32 - 64 type: number epochs: description: >- Passes over the training selection, 1 through 5. The cost estimate grows linearly with epochs. maximum: 5 minimum: 1 type: integer learningRate: description: >- Optimizer learning rate, 0.00001 through 0.001. The default is 0.00015. maximum: 0.001 minimum: 0.00001 type: number loraRank: description: >- LoRA adapter rank: 8, 16, or 32. Defaults to 16. Alpha is twice the rank. enum: - 8 - 16 - 32 type: number maxSequenceLength: description: >- Token budget per training example, prompt and target together: 64 through 16384. The cost estimate is blocked when the longest example exceeds it. Defaults: 12288 for SFT, 3072 for gisting. maximum: 16384 minimum: 64 type: integer method: const: sft description: >- Training method; `sft` learns a LoRA adapter from prompt and target examples. type: string sliceId: description: >- The dataset slice to train on. It must freeze a split field. Training uses its train parts and validation uses its validate parts at the slice cutoff. Rows marked as test by any split field are excluded, except validation rows in a part with both validate and test roles in the selected split field. pattern: >- ^dsl\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string source: $ref: '#/components/schemas/TrainingModelSource' splitFieldId: description: >- The split field to use, identified by field ID. Required when the slice freezes more than one split field. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string trainParts: description: >- The train parts to learn on. Defaults to every part with the train role. items: maxLength: 64 pattern: ^[A-Za-z0-9_]+(?:[.-][A-Za-z0-9_]+)*$ type: string maxItems: 10 minItems: 1 type: array training: $ref: '#/components/schemas/TrainingDatasetSelection' validation: $ref: '#/components/schemas/TrainingDatasetSelection' required: - method - source - loraRank - epochs - learningRate - batchSize - maxSequenceLength - training type: object GistingTrainingConfiguration: additionalProperties: false description: A normalized prompt-gisting Training draft. properties: batchSize: description: >- Sequences per optimizer step across all GPUs: 8, 16, 32, or 64. The default is 8. enum: - 8 - 16 - 32 - 64 type: number epochs: description: >- Passes over the training selection, 1 through 5. The cost estimate grows linearly with epochs. maximum: 5 minimum: 1 type: integer gistTokens: description: >- Number of learned tokens that replace the prompt: 8, 32, 64, or a fraction of the prompt length: "1/4", "1/8", "1/16", or "1/32". Fractions are rounded up to a whole token and limited to 8 through 256 tokens. Defaults to "1/8". enum: - 8 - 32 - 64 - 1/4 - 1/8 - 1/16 - 1/32 learningRate: description: >- Optimizer learning rate for the soft tokens, 0.001 through 0.3. The default is 0.03. maximum: 0.3 minimum: 0.001 type: number maxSequenceLength: description: >- Token budget per training example, prompt and target together: 64 through 16384. The cost estimate is blocked when the longest example exceeds it. Defaults: 12288 for SFT, 3072 for gisting. maximum: 16384 minimum: 64 type: integer method: const: gisting description: >- Training method; `gisting` learns tokens that replace a reusable prompt. type: string promptFileId: description: >- Project file ID with prefix `ppf.` containing the finalized UTF-8 prompt to compress. pattern: >- ^ppf\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string sliceId: description: >- The dataset slice to train on. It must freeze a split field. Training uses its train parts and validation uses its validate parts at the slice cutoff. Rows marked as test by any split field are excluded, except validation rows in a part with both validate and test roles in the selected split field. pattern: >- ^dsl\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string source: $ref: '#/components/schemas/TrainingModelSource' splitFieldId: description: >- The split field to use, identified by field ID. Required when the slice freezes more than one split field. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string trainParts: description: >- The train parts to learn on. Defaults to every part with the train role. items: maxLength: 64 pattern: ^[A-Za-z0-9_]+(?:[.-][A-Za-z0-9_]+)*$ type: string maxItems: 10 minItems: 1 type: array training: $ref: '#/components/schemas/TrainingDatasetSelection' validation: $ref: '#/components/schemas/TrainingDatasetSelection' required: - method - source - promptFileId - gistTokens - epochs - learningRate - batchSize - maxSequenceLength - training type: object DatasetSplitPartExclusions: description: Split-part selections to exclude; omit when no parts should be excluded. items: additionalProperties: false description: Split field and selected part names. properties: fieldId: description: Split field ID with prefix `df.`. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string names: description: Names of the split parts to select. items: maxLength: 64 pattern: ^[A-Za-z0-9_]+(?:[.-][A-Za-z0-9_]+)*$ type: string maxItems: 10 minItems: 1 type: array required: - fieldId - names type: object maxItems: 500 type: array DatasetFieldSourceConditions: description: >- Field-author conditions that selected rows must satisfy; defaults to no conditions. items: additionalProperties: false description: Required authors for a field. properties: fieldId: description: Dataset field ID with prefix `df.`. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string sources: description: Revision authors to match for this field. items: $ref: '#/components/schemas/FieldRevisionSource' minItems: 1 type: array required: - fieldId - sources type: object type: array DatasetFieldValueConditions: description: >- Field-value conditions that selected rows must satisfy; defaults to no conditions. items: additionalProperties: false description: Condition applied to a field value. properties: fieldId: description: Dataset field ID with prefix `df.` to compare. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string jsonPath: description: JSONPath locating the value within the field. minLength: 1 type: string latestOnly: default: true description: >- Whether to test only the latest revision from the allowed authors; false tests any revision, and the default is true. type: boolean operator: description: >- Text comparison: `text_eq` for equality or `text_neq` for inequality. enum: - text_eq - text_neq type: string sources: default: [] description: Allowed revision authors; an empty array allows any author. items: $ref: '#/components/schemas/FieldRevisionSource' type: array value: description: Scalar value compared as text; null is treated as an empty string. type: - string - number - boolean - 'null' required: - jsonPath - operator - value - fieldId type: object type: array DatasetSplitPartInclusion: additionalProperties: false description: Split parts to include; omit to include every part. properties: fieldId: description: Split field ID with prefix `df.`. pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string names: description: Names of the split parts to select. items: maxLength: 64 pattern: ^[A-Za-z0-9_]+(?:[.-][A-Za-z0-9_]+)*$ type: string maxItems: 10 minItems: 1 type: array required: - fieldId - names type: object DatasetSliceSelection: additionalProperties: false description: >- Slice whose frozen rows and fields to read; omit to use live dataset rows. properties: sliceId: description: Dataset slice ID with prefix `dsl.`. pattern: >- ^dsl\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string required: - sliceId type: object TrainingInputExportFailure: anyOf: - properties: code: description: >- Failure category: `empty_selection` when no rows match, `size_limit_exceeded` when the export is too large, or `assembly_failed` when preparation fails. enum: - empty_selection - size_limit_exceeded - assembly_failed type: string message: description: Explanation of why the export failed. maxLength: 500 minLength: 1 type: string timestamp: description: Time the export failed, as an ISO 8601 timestamp. format: date-time pattern: >- ^(?:(?:\d\d[2468][048]|\d\d[13579][26]|\d\d0[48]|[02468][048]00|[13579][26]00)-02-29|\d{4}-(?:(?:0[13578]|1[02])-(?:0[1-9]|[12]\d|3[01])|(?:0[469]|11)-(?:0[1-9]|[12]\d|30)|(?:02)-(?:0[1-9]|1\d|2[0-8])))T(?:(?:[01]\d|2[0-3]):[0-5]\d:[0-5]\d(?:\.\d+)?(?:Z))$ type: string required: - timestamp - code - message type: object - type: 'null' description: Export failure details, or null when the export did not fail. TrainingModelSource: description: The exact Base Model or Model Version used as the source. oneOf: - additionalProperties: false properties: baseModelId: description: Catalog base model identifier. enum: - qwen3-4b - qwen3-8b - qwen3-32b - qwen3-235b type: string type: const: base_model description: >- Training source: `base_model` uses a catalog model; `model_version` uses a published project model version. type: string required: - type - baseModelId type: object - additionalProperties: false properties: modelVersionId: description: Source model version ID with prefix `mdv.`. pattern: >- ^mdv\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string type: const: model_version description: >- Training source: `base_model` uses a catalog model; `model_version` uses a published project model version. type: string required: - type - modelVersionId type: object TrainingDatasetSelection: additionalProperties: false description: A Dataset and normalized filter selected for one Training input purpose. properties: datasetId: description: Dataset ID with prefix `ds.` containing the selected examples. pattern: >- ^ds\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string itemFilter: $ref: '#/components/schemas/TrainingDatasetItemFilter' description: >- Conditions selecting dataset rows; omitted conditions do not restrict the selection. required: - datasetId - itemFilter type: object FieldRevisionSource: description: Source that wrote a field revision. oneOf: - additionalProperties: false description: Source that wrote a field revision. properties: type: const: user description: >- Revision author: `user` for a user action or `system` for a platform operation. type: string userId: description: Identifier of the user who wrote the revision. minLength: 1 type: string required: - type - userId type: object - additionalProperties: false description: Source that wrote a field revision. properties: source: description: Identifier of the platform operation that wrote the revision. minLength: 1 type: string type: const: system description: >- Revision author: `user` for a user action or `system` for a platform operation. type: string required: - type - source type: object TrainingDatasetItemFilter: additionalProperties: false description: A normalized Dataset item filter stored with a Training draft. properties: excludeParts: $ref: '#/components/schemas/DatasetSplitPartExclusions' fieldSources: $ref: '#/components/schemas/DatasetFieldSourceConditions' fieldValues: $ref: '#/components/schemas/DatasetFieldValueConditions' missingRevisionFields: description: >- Field IDs with prefix `df.` that must have no revision on a selected row; defaults to an empty array. items: pattern: >- ^df\.[0-9a-f]{8}-[0-9a-f]{4}-[7][0-9a-f]{3}-[89ab][0-9a-f]{3}-[0-9a-f]{12}$ type: string type: array parts: $ref: '#/components/schemas/DatasetSplitPartInclusion' slice: $ref: '#/components/schemas/DatasetSliceSelection' required: - fieldValues - fieldSources - missingRevisionFields type: object headers: RateLimit-Limit: description: The maximum number of requests allowed in the current window. schema: type: integer RateLimit-Policy: description: The rate limit policy (e.g., "60;w=60"). schema: type: string RateLimit-Remaining: description: The number of requests remaining in the current window. schema: type: integer RateLimit-Reset: description: The number of seconds until the current window resets. schema: type: integer X-Request-Id: description: >- The request id: the X-Request-Id you sent when it is 1-128 characters of letters, digits, ".", "_" or "-", otherwise a generated UUID. Present on every response, including errors; quote it when reporting a problem. schema: type: string securitySchemes: apiKeyAuth: description: >- Enter `ApiKey ` in Authorization, including the `ApiKey ` prefix. In cURL and Python samples, `` means this full value. The TypeScript SDK adds the prefix for you. in: header name: Authorization type: apiKey scopedTokenAuth: description: >- Enter the project-scoped token as `pst_...` in X-Scoped-Token, without an `ApiKey ` prefix. Create a token with POST /projects/{projectId}/tokens. in: header name: X-Scoped-Token type: apiKey ```` This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.