runbios-sdk 0.2.1-dev.111 → 0.2.1-dev.113

package/README.md

@@ -239,17 +239,47 @@ const preview = await client.datasets.preview('ds_abc123', { pageSize: 5 });
 
 // Import from HuggingFace
 const imported = await client.datasets.importFromHuggingFace({
-  repoId: 'databricks/dolly-15k',
+  repoId: 'HuggingFaceH4/ultrachat_200k',
   integrationId: 'int_abc123',
-  name: 'Dolly 15k',
+  name: 'Ultrachat SFT',
+  subset: 'default',
+  split: 'train_sft',
 });
 
 // Validate before upload
 const validation = await client.datasets.validate('./data.jsonl');
 ```
 
+#### Connected Hub search and exact source selection
+
+Use `client.integrations.list()` to discover connected accounts, then select the
+returned ID explicitly. `client.integrations.browse(integrationId, { query,
+page, limit })` searches with that account's access. Discovery returns account
+metadata, never stored credentials.
+
+Pass the same `integrationId` to `datasets.previewHub` and
+`datasets.importFromHuggingFace`. Preview returns `available_configs` and
+`available_splits`; choose the actual subset/split rather than assuming `train`.
+A missing or unauthorized integration is an error, not anonymous fallback.
+Public imports without an integration use `datasets.registerHuggingFace` and
+require `workspaceId` on the client or method call.
+
+Both import methods accept `revision`, `maxSamples`, `sampleStrategy` and
+`importMode`. The service records immutable source metadata for resume. A Hub
+preview does not prove readiness or preview an arbitrary older revision: inspect
+the registered dataset when importing an explicit revision. Poll
+`datasets.getStatus(id)` until `ready` before selecting it for training; a client
+polling timeout does not cancel the background import.
+
 ### Training
 
+`datasetIds` preserves source order. Pass the same `mixing` object to preflight
+and create: `{ mode: 'interleave', weights: [3, 1], seed: 42 }`. Weights match the
+ordered IDs. Modes are `sequential`, `shuffle`, `interleave`, and `phased`; phased
+plans contain ordered `phases` with `name`, `portion`, optional `weights`, and
+`shuffle`. The default composition seed is 42. Resume reconstructs data from
+source pins and the stored plan and verifies the checksum.
+
 ```typescript
 const request = {
   idempotencyKey: 'training-create-20260711-0001',

package/dist/index.d.ts

@@ -26,6 +26,7 @@
 import type { BiOSConfig, ApiKeyIntrospection } from './types.js';
 import { Models } from './resources/models.js';
 import { Datasets } from './resources/datasets.js';
+import { Integrations } from './resources/integrations.js';
 import { Training } from './resources/training.js';
 import { Wallet } from './resources/wallet.js';
 import { GPU } from './resources/gpu.js';
@@ -35,12 +36,13 @@ import { Loop } from './resources/loop.js';
  * SDK version. Sent as part of the User-Agent header.
  * Must match package.json "version" -- enforced by a contract test.
  */
-export declare const VERSION = "0.2.1-dev.111";
+export declare const VERSION = "0.2.1-dev.113";
 export declare class RunBiOS {
     /** Search models, fetch configs, check adapter compatibility. */
     readonly models: Models;
     /** Upload, import, preview, and manage training datasets. */
     readonly datasets: Datasets;
+    readonly integrations: Integrations;
     /** Create, monitor, stop, and resume fine-tuning jobs. */
     readonly training: Training;
     /** Check wallet balance. */
@@ -68,6 +70,7 @@ export { RunBiOS as BiOS };
 export { ApiError, GpuRejectionError, CapacityUnavailableError, ComingSoonError, CAPACITY_UNAVAILABLE_CODE, COMING_SOON_CODES, PERMANENT_GPU_CODES, isPermanentGpuCode, gpuRejectionCodeForReason, } from './client.js';
 export { Models } from './resources/models.js';
 export { Datasets } from './resources/datasets.js';
+export { Integrations, type HuggingFaceIntegration, type IntegrationBrowseParams } from './resources/integrations.js';
 export { Training } from './resources/training.js';
 export { Wallet } from './resources/wallet.js';
 export { GPU, type GPURecommendation } from './resources/gpu.js';

package/dist/index.js

@@ -26,6 +26,7 @@
 import { HttpClient, envInferenceKey } from './client.js';
 import { Models } from './resources/models.js';
 import { Datasets } from './resources/datasets.js';
+import { Integrations } from './resources/integrations.js';
 import { Training } from './resources/training.js';
 import { Wallet } from './resources/wallet.js';
 import { GPU } from './resources/gpu.js';
@@ -35,12 +36,13 @@ import { Loop } from './resources/loop.js';
  * SDK version. Sent as part of the User-Agent header.
  * Must match package.json "version" -- enforced by a contract test.
  */
-export const VERSION = '0.2.1-dev.111';
+export const VERSION = '0.2.1-dev.113';
 export class RunBiOS {
     /** Search models, fetch configs, check adapter compatibility. */
     models;
     /** Upload, import, preview, and manage training datasets. */
     datasets;
+    integrations;
     /** Create, monitor, stop, and resume fine-tuning jobs. */
     training;
     /** Check wallet balance. */
@@ -65,6 +67,7 @@ export class RunBiOS {
         this._http = http;
         this.models = new Models(http);
         this.datasets = new Datasets(http);
+        this.integrations = new Integrations(http);
         this.training = new Training(http);
         this.wallet = new Wallet(http);
         this.gpu = new GPU(http);
@@ -95,6 +98,7 @@ export { RunBiOS as BiOS };
 export { ApiError, GpuRejectionError, CapacityUnavailableError, ComingSoonError, CAPACITY_UNAVAILABLE_CODE, COMING_SOON_CODES, PERMANENT_GPU_CODES, isPermanentGpuCode, gpuRejectionCodeForReason, } from './client.js';
 export { Models } from './resources/models.js';
 export { Datasets } from './resources/datasets.js';
+export { Integrations } from './resources/integrations.js';
 export { Training } from './resources/training.js';
 export { Wallet } from './resources/wallet.js';
 export { GPU } from './resources/gpu.js';

package/dist/resources/datasets.js

@@ -203,6 +203,7 @@ export class Datasets {
         return normalizeDataset(await this._http.fetchPost(`/api/datasets/integrations/${encodeURIComponent(params.integrationId)}/import`, {
             dataset_id: params.repoId,
             name: params.name || params.repoId.split('/').pop() || params.repoId,
+            revision: params.revision,
             subset: params.subset,
             split: params.split,
             max_samples: params.maxSamples,
@@ -239,6 +240,7 @@ export class Datasets {
             description: raw.description,
             hf_subset: raw.subset ?? raw.hf_subset,
             hf_split: raw.split ?? raw.hf_split,
+            hf_revision: raw.revision ?? raw.hf_revision,
             max_samples: raw.maxSamples ?? raw.max_samples,
             sample_strategy: raw.sampleStrategy ?? raw.sample_strategy,
             column_mapping: raw.columnMapping ?? raw.column_mapping,
@@ -348,6 +350,8 @@ export class Datasets {
             q.set('subset', params.subset);
         if (params.limit !== undefined)
             q.set('limit', String(params.limit));
+        if (params.integrationId)
+            q.set('integration_id', params.integrationId);
         return this._http.fetchGet(`/api/datasets/hub-preview?${q}`);
     }
     /**

package/dist/resources/integrations.d.ts

@@ -0,0 +1,21 @@
+import type { HttpClient } from '../client.js';
+export interface HuggingFaceIntegration {
+    id: string;
+    label: string;
+    provider: string;
+    status: string;
+    hf_username?: string;
+    scope?: string;
+}
+export interface IntegrationBrowseParams {
+    query?: string;
+    page?: number;
+    limit?: number;
+    sort?: 'downloads' | 'likes' | 'modified';
+}
+export declare class Integrations {
+    private readonly _http;
+    constructor(_http: HttpClient);
+    list(): Promise<HuggingFaceIntegration[]>;
+    browse(id: string, params?: IntegrationBrowseParams): Promise<Record<string, unknown>>;
+}

package/dist/resources/integrations.js

@@ -0,0 +1,20 @@
+export class Integrations {
+    _http;
+    constructor(_http) {
+        this._http = _http;
+    }
+    async list() {
+        const result = await this._http.fetchGet('/api/datasets/integrations');
+        if (!Array.isArray(result.integrations))
+            throw new Error('The integration service did not return an account list');
+        return result.integrations;
+    }
+    async browse(id, params = {}) {
+        const query = new URLSearchParams({ page: String(params.page ?? 1), limit: String(params.limit ?? 20) });
+        if (params.query !== undefined)
+            query.set('search', params.query);
+        if (params.sort !== undefined)
+            query.set('sort', params.sort);
+        return this._http.fetchGet(`/api/datasets/integrations/${encodeURIComponent(id)}/browse?${query}`);
+    }
+}

package/dist/resources/loop.d.ts

@@ -187,7 +187,11 @@ export declare class Loop {
      * - `sft` -- answers you marked right, and answers you rewrote.
      * - `dpo` -- answers you REWROTE, so a better and a worse version of the same
      *   reply exist. Nothing else produces a pair.
-     * - `grpo` -- answers with a value or fact they can be checked against.
+     * - `grpo` -- answers with a value or fact they can be checked against, or
+     *   a judge rubric, which scores answers the run has not written yet.
+     * - `kto` -- anything carrying a yes or a no, INCLUDING a thumbs-down with
+     *   nothing written. That row trains nothing under the other three methods,
+     *   which is why this one exists: it is the feedback people actually give.
      *
      * `holdout_percent` holds back your most recent work rather than a random
      * slice, so the evaluation measures whether the model generalised instead of

package/dist/resources/loop.js

@@ -240,7 +240,11 @@ export class Loop {
      * - `sft` -- answers you marked right, and answers you rewrote.
      * - `dpo` -- answers you REWROTE, so a better and a worse version of the same
      *   reply exist. Nothing else produces a pair.
-     * - `grpo` -- answers with a value or fact they can be checked against.
+     * - `grpo` -- answers with a value or fact they can be checked against, or
+     *   a judge rubric, which scores answers the run has not written yet.
+     * - `kto` -- anything carrying a yes or a no, INCLUDING a thumbs-down with
+     *   nothing written. That row trains nothing under the other three methods,
+     *   which is why this one exists: it is the feedback people actually give.
      *
      * `holdout_percent` holds back your most recent work rather than a random
      * slice, so the evaluation measures whether the model generalised instead of

package/dist/types.d.ts

@@ -396,8 +396,9 @@ export interface DatasetPreview {
 }
 /** Parameters for importing a dataset from HuggingFace Hub. */
 export interface DatasetImportHFParams {
-    /** HuggingFace dataset repository ID (e.g. "databricks/dolly-15k"). */
+    /** HuggingFace dataset repository ID (e.g. "HuggingFaceH4/ultrachat_200k"). */
     repoId: string;
+    revision?: string;
     /** Display name for the imported dataset. */
     name?: string;
     /** Dataset subset/config to import. */
@@ -418,6 +419,7 @@ export interface DatasetImportHFParams {
 /** Parameters for registering a HuggingFace dataset without an integration. */
 export interface DatasetRegisterHFParams {
     repoId: string;
+    revision?: string;
     name?: string;
     workspaceId?: string;
     description?: string;
@@ -441,6 +443,7 @@ export interface DatasetHubSearchParams {
 export interface DatasetHubPreviewParams {
     /** HuggingFace dataset ID. */
     datasetId: string;
+    integrationId?: string;
     /** Split to preview. Defaults to "train". */
     split?: string;
     /** Subset/config name. */
@@ -1956,7 +1959,7 @@ export interface ChatCompletionUsage {
     prompt_tokens_details?: PromptTokensDetails;
 }
 /** What a training set is shaped for. The three need different things. */
-export type LoopMethod = 'sft' | 'dpo' | 'grpo';
+export type LoopMethod = 'sft' | 'dpo' | 'grpo' | 'kto';
 /** Who judged an answer. */
 export type LoopSource = 'human' | 'verifier' | 'judge' | 'behavioural';
 /** The judgement itself. */

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "runbios-sdk",
-  "version": "0.2.1-dev.111",
+  "version": "0.2.1-dev.113",
   "description": "Official TypeScript SDK for the Run BiOS training and deployment platform API",
   "type": "module",
   "main": "./dist/index.js",