opik 2.0.0 → 2.0.1

package/dist/index.d.cts

@@ -12503,9 +12503,10 @@ declare class Dataset<T extends DatasetItemData = DatasetItemData> {
      * Retrieve raw DatasetItem objects with full metadata (evaluators, executionPolicy) preserved.
      *
      * @param nbSamples The number of samples to retrieve. If not set - all items are returned
+     * @param lastRetrievedId Optional ID of the last retrieved item for pagination
      * @returns A list of DatasetItem objects
      */
-    getRawItems(nbSamples?: number): Promise<DatasetItem<T>[]>;
+    getRawItems(nbSamples?: number, lastRetrievedId?: string): Promise<DatasetItem<T>[]>;
     /**
      * Insert items from a JSON string array into the dataset.
      *
@@ -12763,6 +12764,10 @@ interface EvaluateTestSuiteOptions<T = Record<string, unknown>> {
     client?: OpikClient;
     /** Optional list of tags to associate with the experiment */
     tags?: string[];
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
+    /** Limit the number of dataset items to evaluate. If not set, all items are evaluated. */
+    nbSamples?: number;
 }
 /**
  * Run a test suite using evaluators and execution policy stored in the dataset version metadata.
@@ -12783,7 +12788,6 @@ interface CreateTestSuiteOptions {
 interface UpdateTestSuiteOptions {
     globalAssertions?: string[];
     globalExecutionPolicy?: ExecutionPolicy;
-    tags?: string[];
 }
 declare class TestSuite {
     private readonly _dataset;
@@ -12820,7 +12824,7 @@ declare class TestSuite {
      * @throws Error if any item is missing an `id`
      */
     update(items: UpdateTestSuiteItem[]): Promise<void>;
-    getItems(): Promise<Array<{
+    getItems(nbSamples?: number, lastRetrievedId?: string): Promise<Array<{
         id: string;
         data: Record<string, unknown>;
         description?: string;
@@ -12889,6 +12893,10 @@ interface RunTestsOptions {
     experimentTags?: string[];
     /** Optional model name override for LLMJudge evaluators */
     model?: string;
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
+    /** Limit the number of dataset items to evaluate. If not set, all items are evaluated. */
+    nbSamples?: number;
 }
 /**
  * Run a test suite evaluation against a task function.
@@ -14118,6 +14126,8 @@ interface EvaluateOptions<T = Record<string, unknown>> {
     scoringKeyMapping?: ScoringKeyMappingType;
     /** Optional list of tags to associate with the experiment */
     tags?: string[];
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
     /** Optional agent configuration blueprint ID to link with the experiment */
     blueprintId?: string;
 }

package/dist/index.d.ts

@@ -12503,9 +12503,10 @@ declare class Dataset<T extends DatasetItemData = DatasetItemData> {
      * Retrieve raw DatasetItem objects with full metadata (evaluators, executionPolicy) preserved.
      *
      * @param nbSamples The number of samples to retrieve. If not set - all items are returned
+     * @param lastRetrievedId Optional ID of the last retrieved item for pagination
      * @returns A list of DatasetItem objects
      */
-    getRawItems(nbSamples?: number): Promise<DatasetItem<T>[]>;
+    getRawItems(nbSamples?: number, lastRetrievedId?: string): Promise<DatasetItem<T>[]>;
     /**
      * Insert items from a JSON string array into the dataset.
      *
@@ -12763,6 +12764,10 @@ interface EvaluateTestSuiteOptions<T = Record<string, unknown>> {
     client?: OpikClient;
     /** Optional list of tags to associate with the experiment */
     tags?: string[];
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
+    /** Limit the number of dataset items to evaluate. If not set, all items are evaluated. */
+    nbSamples?: number;
 }
 /**
  * Run a test suite using evaluators and execution policy stored in the dataset version metadata.
@@ -12783,7 +12788,6 @@ interface CreateTestSuiteOptions {
 interface UpdateTestSuiteOptions {
     globalAssertions?: string[];
     globalExecutionPolicy?: ExecutionPolicy;
-    tags?: string[];
 }
 declare class TestSuite {
     private readonly _dataset;
@@ -12820,7 +12824,7 @@ declare class TestSuite {
      * @throws Error if any item is missing an `id`
      */
     update(items: UpdateTestSuiteItem[]): Promise<void>;
-    getItems(): Promise<Array<{
+    getItems(nbSamples?: number, lastRetrievedId?: string): Promise<Array<{
         id: string;
         data: Record<string, unknown>;
         description?: string;
@@ -12889,6 +12893,10 @@ interface RunTestsOptions {
     experimentTags?: string[];
     /** Optional model name override for LLMJudge evaluators */
     model?: string;
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
+    /** Limit the number of dataset items to evaluate. If not set, all items are evaluated. */
+    nbSamples?: number;
 }
 /**
  * Run a test suite evaluation against a task function.
@@ -14118,6 +14126,8 @@ interface EvaluateOptions<T = Record<string, unknown>> {
     scoringKeyMapping?: ScoringKeyMappingType;
     /** Optional list of tags to associate with the experiment */
     tags?: string[];
+    /** Number of concurrent task executions (default: 16, matching Python SDK) */
+    taskThreads?: number;
     /** Optional agent configuration blueprint ID to link with the experiment */
     blueprintId?: string;
 }

package/dist/index.js

@@ -1 +1 @@
-export{oa as AgentTaskCompletionJudge,na as AgentToolCorrectnessJudge,ba as AnswerRelevance,Z as BaseLLMJudgeMetric,y as BaseMetric,z as BaseSuiteEvaluator,m as ChatPrompt,qa as ComplianceRiskJudge,i as ConfigMismatchError,h as ConfigNotFoundError,W as Contains,v as DEFAULT_EXECUTION_POLICY,j as Dataset,f as DatasetVersion,g as DatasetVersionNotFoundError,ia as DemographicBiasJudge,ga as DialogueHelpfulnessJudge,V as ExactMatch,ca as GEval,da as GEvalPreset,ka as GenderBiasJudge,aa as Hallucination,Y as IsJson,M as LLMJudge,D as ModelConfigurationError,B as ModelError,C as ModelGenerationError,_ as Moderation,va as Opik,A as OpikBaseModel,n as OpikQueryLanguage,d as OpikSpanType,ja as PoliticalBiasJudge,l as Prompt,k as PromptType,pa as PromptUncertaintyJudge,ha as QARelevanceJudge,X as RegexMatch,ma as RegionalBiasJudge,la as ReligiousBiasJudge,L as ResponseSchema,J as SYSTEM_PROMPT,fa as SummarizationCoherenceJudge,ea as SummarizationConsistencyJudge,ua as TestSuite,w as TestSuiteResult,p as ThreadsAnnotationQueue,o as TracesAnnotationQueue,K as USER_PROMPT_TEMPLATE,$ as Usefulness,F as VercelAIChatModel,s as activateRunner,q as agentConfigContext,x as buildSuiteResult,G as createModel,H as createModelFromInstance,O as deserializeEvaluators,E as detectProvider,c as disableLogger,T as evaluate,U as evaluatePrompt,R as evaluateTestSuite,r as flushAll,e as generateId,t as getTrackContext,a as logger,ra as resolveEvaluators,P as resolveExecutionPolicy,Q as resolveItemExecutionPolicy,I as resolveModel,S as runTests,N as serializeEvaluators,b as setLoggerLevel,u as track,sa as validateEvaluators,ta as validateExecutionPolicy,wa as z}from'./chunk-2AOEXUQ4.js';
+export{oa as AgentTaskCompletionJudge,na as AgentToolCorrectnessJudge,ba as AnswerRelevance,Z as BaseLLMJudgeMetric,y as BaseMetric,z as BaseSuiteEvaluator,m as ChatPrompt,qa as ComplianceRiskJudge,i as ConfigMismatchError,h as ConfigNotFoundError,W as Contains,v as DEFAULT_EXECUTION_POLICY,j as Dataset,f as DatasetVersion,g as DatasetVersionNotFoundError,ia as DemographicBiasJudge,ga as DialogueHelpfulnessJudge,V as ExactMatch,ca as GEval,da as GEvalPreset,ka as GenderBiasJudge,aa as Hallucination,Y as IsJson,M as LLMJudge,D as ModelConfigurationError,B as ModelError,C as ModelGenerationError,_ as Moderation,va as Opik,A as OpikBaseModel,n as OpikQueryLanguage,d as OpikSpanType,ja as PoliticalBiasJudge,l as Prompt,k as PromptType,pa as PromptUncertaintyJudge,ha as QARelevanceJudge,X as RegexMatch,ma as RegionalBiasJudge,la as ReligiousBiasJudge,L as ResponseSchema,J as SYSTEM_PROMPT,fa as SummarizationCoherenceJudge,ea as SummarizationConsistencyJudge,ua as TestSuite,w as TestSuiteResult,p as ThreadsAnnotationQueue,o as TracesAnnotationQueue,K as USER_PROMPT_TEMPLATE,$ as Usefulness,F as VercelAIChatModel,s as activateRunner,q as agentConfigContext,x as buildSuiteResult,G as createModel,H as createModelFromInstance,O as deserializeEvaluators,E as detectProvider,c as disableLogger,T as evaluate,U as evaluatePrompt,R as evaluateTestSuite,r as flushAll,e as generateId,t as getTrackContext,a as logger,ra as resolveEvaluators,P as resolveExecutionPolicy,Q as resolveItemExecutionPolicy,I as resolveModel,S as runTests,N as serializeEvaluators,b as setLoggerLevel,u as track,sa as validateEvaluators,ta as validateExecutionPolicy,wa as z}from'./chunk-HRNPUK4B.js';

package/dist/{suite-DE3AOLJG.js → suite-RWWP3MWS.js}

@@ -1 +1 @@
-export{v as DEFAULT_EXECUTION_POLICY,ua as TestSuite,w as TestSuiteResult,x as buildSuiteResult,O as deserializeEvaluators,R as evaluateTestSuite,P as resolveExecutionPolicy,Q as resolveItemExecutionPolicy,S as runTests,N as serializeEvaluators}from'./chunk-2AOEXUQ4.js';
+export{v as DEFAULT_EXECUTION_POLICY,ua as TestSuite,w as TestSuiteResult,x as buildSuiteResult,O as deserializeEvaluators,R as evaluateTestSuite,P as resolveExecutionPolicy,Q as resolveItemExecutionPolicy,S as runTests,N as serializeEvaluators}from'./chunk-HRNPUK4B.js';

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "opik",
   "description": "Opik TypeScript and JavaScript SDK",
-  "version": "2.0.0",
+  "version": "2.0.1",
   "repository": {
     "type": "git",
     "url": "git+https://github.com/comet-ml/opik.git",