opik 2.0.7 → 2.0.9

package/dist/index.d.cts

@@ -12672,6 +12672,28 @@ type EvaluationTestResult = {
  * Mirrors the Fern-generated ExecutionPolicyWrite type.
  */
 type ExecutionPolicy = ExecutionPolicyWrite;
+/**
+ * A raw test suite item as returned by {@link TestSuite.getRawItems}.
+ *
+ * Unlike the view returned by {@link TestSuite.getItems}, this preserves:
+ * - `evaluators` as raw {@link EvaluatorItemWrite} objects (not decoded to assertion strings)
+ * - `executionPolicy` as the item-level value only (not merged with the suite-level default)
+ *
+ * `data` is the full stored payload exactly as returned by the dataset
+ * (so if an item was stored with a `description`, it will be present in
+ * `data` as well). `description` is additionally exposed as a top-level
+ * field for ergonomic access.
+ *
+ * Use this when you need to introspect or forward the stored evaluator
+ * config or per-item execution policy verbatim.
+ */
+interface RawTestSuiteItem {
+    id: string;
+    data: DatasetItemData;
+    description?: string;
+    evaluators?: EvaluatorItemWrite[];
+    executionPolicy?: ExecutionPolicy;
+}
 declare const DEFAULT_EXECUTION_POLICY: Required<ExecutionPolicy>;
 /**
  * A single item to be inserted into a test suite via `insert()`.
@@ -12854,6 +12876,24 @@ declare class TestSuite {
         assertions: string[];
         executionPolicy: Required<ExecutionPolicy>;
     }>>;
+    /**
+     * Retrieve items with full suite-specific metadata preserved verbatim.
+     *
+     * Unlike {@link getItems}, this does NOT:
+     * - Decode `evaluators` into assertion strings (you get raw {@link EvaluatorItemWrite}[])
+     * - Merge the item-level `executionPolicy` with the suite-level default
+     *
+     * `data` mirrors the stored payload (includes `description` if present);
+     * `description` is additionally exposed at the top level for convenience.
+     *
+     * Use this when you need to inspect or forward the stored evaluator
+     * config or per-item execution policy as-is.
+     *
+     * @param nbSamples Max items to retrieve. Omit to return all items.
+     * @param lastRetrievedId Opaque cursor for pagination (last `id` from a previous call).
+     * @returns Array of {@link RawTestSuiteItem} objects.
+     */
+    getRawItems(nbSamples?: number, lastRetrievedId?: string): Promise<RawTestSuiteItem[]>;
     getGlobalAssertions(): Promise<string[]>;
     getTags(): Promise<string[]>;
     getItemsCount(): Promise<number | undefined>;
@@ -13151,6 +13191,25 @@ declare class Experiment {
     getUrl(): Promise<string>;
 }
 
+interface TestSuiteExperimentData extends ExperimentData {
+    passRate?: number;
+    passedCount?: number;
+    totalCount?: number;
+    assertionScores?: AssertionScoreAveragePublic[];
+}
+/**
+ * Represents an experiment run against a test suite. Extends `Experiment`
+ * with the aggregate assertion statistics the backend populates only for
+ * evaluation-suite experiments (null/undefined for regular dataset experiments).
+ */
+declare class TestSuiteExperiment extends Experiment {
+    readonly passRate?: number;
+    readonly passedCount?: number;
+    readonly totalCount?: number;
+    readonly assertionScores?: AssertionScoreAveragePublic[];
+    constructor(data: TestSuiteExperimentData, opik: OpikClient);
+}
+
 interface ChatPromptData extends BasePromptData {
     messages: ChatMessage[];
 }
@@ -13596,6 +13655,25 @@ declare class OpikClient {
      * @throws {DatasetNotFoundError} If the dataset doesn't exist
      */
     getDatasetExperiments: (datasetName: string, maxResults?: number, projectName?: string) => Promise<Experiment[]>;
+    /**
+     * Retrieves all experiments associated with a test suite.
+     *
+     * @param name The name of the test suite
+     * @param maxResults Maximum number of experiments to return (default: 100)
+     * @param projectName Optional project name to scope the suite lookup. If not provided, uses the client's configured project.
+     * @returns A list of TestSuiteExperiment objects associated with the test suite,
+     *   each carrying the suite-specific assertion aggregates (`passRate`, `passedCount`,
+     *   `totalCount`, `assertionScores`) populated by the backend.
+     * @throws {DatasetNotFoundError} If the test suite doesn't exist
+     */
+    getTestSuiteExperiments: (name: string, maxResults?: number, projectName?: string) => Promise<TestSuiteExperiment[]>;
+    /**
+     * Paginated fetch of experiments for a given dataset ID, mapping each raw
+     * `ExperimentPublic` row to a caller-chosen entity. Used internally by
+     * `getDatasetExperiments` and `getTestSuiteExperiments` to share the
+     * loop shape and only differ on the constructed type.
+     */
+    private findExperimentsByDatasetId;
     /**
      * Deletes an experiment by ID
      *
@@ -16013,4 +16091,4 @@ declare class ConfigMismatchError extends OpikError {
 
 declare function activateRunner(): void;
 
-export { AgentTaskCompletionJudge, AgentToolCorrectnessJudge, type AllProviderOptions, AnnotationQueuePublicScope as AnnotationQueueScope, AnswerRelevance, type AnthropicProviderOptions, BaseLLMJudgeMetric, BaseMetric, BaseSuiteEvaluator, ChatPrompt, ComplianceRiskJudge, type Config, ConfigMismatchError, ConfigNotFoundError, Contains, type CreateTestSuiteOptions, DEFAULT_EXECUTION_POLICY, Dataset, type DatasetPublic, DatasetVersion, DatasetVersionNotFoundError, type DatasetVersionPublic, DemographicBiasJudge, DialogueHelpfulnessJudge, type ErrorInfo, type EvaluateOptions, type EvaluatePromptOptions, type EvaluateTestSuiteOptions, type EvaluationError, type EvaluationResult, type EvaluationScoreResult, type EvaluationTask, type EvaluationTestCase, type EvaluationTestResult, ExactMatch, type ExecutionPolicy, type FeedbackScoreData, type FewShotExampleAnswerRelevanceNoContext, type FewShotExampleAnswerRelevanceWithContext, type FewShotExampleHallucination, type FewShotExampleModeration, type FilterExpression, GEval, GEvalPreset, GenderBiasJudge, type GoogleProviderOptions, Hallucination, IsJson, type ItemResult, LLMJudge, type LLMJudgeConfig, type LLMJudgeModelSettings, type LLMJudgeOptions, type LLMJudgeResponseFormat, ModelConfigurationError, ModelError, ModelGenerationError, Moderation, type OpenAIProviderOptions, OpikClient as Opik, type OpikAssistantMessage, OpikBaseModel, type OpikConfig, type OpikMessage, OpikQueryLanguage, SpanType as OpikSpanType, type OpikSystemMessage, type OpikToolMessage, type OpikUserMessage, type Param, PoliticalBiasJudge, Prompt, PromptType, PromptUncertaintyJudge, type ProviderOptionsForModel, QARelevanceJudge, RegexMatch, RegionalBiasJudge, type RegistryEntry, ReligiousBiasJudge, ResponseSchema, type RunTestsOptions, SYSTEM_PROMPT, type ScoringKeyMappingType, Span, SpanType, SummarizationCoherenceJudge, SummarizationConsistencyJudge, type SupportedModelId, TestSuite, type TestSuiteItem, TestSuiteResult, ThreadsAnnotationQueue, Trace, TracesAnnotationQueue, USER_PROMPT_TEMPLATE, type UpdateTestSuiteItem, type UpdateTestSuiteOptions, Usefulness, VercelAIChatModel, activateRunner, agentConfigContext, buildSuiteResult, createModel, createModelFromInstance, deserializeEvaluators, detectProvider, disableLogger, evaluate, evaluatePrompt, evaluateTestSuite, flushAll, generateId, getGlobalClient, getTrackContext, logger, resetGlobalClient, resolveEvaluators, resolveExecutionPolicy, resolveItemExecutionPolicy, resolveModel, runTests, serializeEvaluators, setGlobalClient, setLoggerLevel, track, validateEvaluators, validateExecutionPolicy };
+export { AgentTaskCompletionJudge, AgentToolCorrectnessJudge, type AllProviderOptions, AnnotationQueuePublicScope as AnnotationQueueScope, AnswerRelevance, type AnthropicProviderOptions, BaseLLMJudgeMetric, BaseMetric, BaseSuiteEvaluator, ChatPrompt, ComplianceRiskJudge, type Config, ConfigMismatchError, ConfigNotFoundError, Contains, type CreateTestSuiteOptions, DEFAULT_EXECUTION_POLICY, Dataset, type DatasetPublic, DatasetVersion, DatasetVersionNotFoundError, type DatasetVersionPublic, DemographicBiasJudge, DialogueHelpfulnessJudge, type ErrorInfo, type EvaluateOptions, type EvaluatePromptOptions, type EvaluateTestSuiteOptions, type EvaluationError, type EvaluationResult, type EvaluationScoreResult, type EvaluationTask, type EvaluationTestCase, type EvaluationTestResult, ExactMatch, type ExecutionPolicy, type FeedbackScoreData, type FewShotExampleAnswerRelevanceNoContext, type FewShotExampleAnswerRelevanceWithContext, type FewShotExampleHallucination, type FewShotExampleModeration, type FilterExpression, GEval, GEvalPreset, GenderBiasJudge, type GoogleProviderOptions, Hallucination, IsJson, type ItemResult, LLMJudge, type LLMJudgeConfig, type LLMJudgeModelSettings, type LLMJudgeOptions, type LLMJudgeResponseFormat, ModelConfigurationError, ModelError, ModelGenerationError, Moderation, type OpenAIProviderOptions, OpikClient as Opik, type OpikAssistantMessage, OpikBaseModel, type OpikConfig, type OpikMessage, OpikQueryLanguage, SpanType as OpikSpanType, type OpikSystemMessage, type OpikToolMessage, type OpikUserMessage, type Param, PoliticalBiasJudge, Prompt, PromptType, PromptUncertaintyJudge, type ProviderOptionsForModel, QARelevanceJudge, type RawTestSuiteItem, RegexMatch, RegionalBiasJudge, type RegistryEntry, ReligiousBiasJudge, ResponseSchema, type RunTestsOptions, SYSTEM_PROMPT, type ScoringKeyMappingType, Span, SpanType, SummarizationCoherenceJudge, SummarizationConsistencyJudge, type SupportedModelId, TestSuite, type TestSuiteItem, TestSuiteResult, ThreadsAnnotationQueue, Trace, TracesAnnotationQueue, USER_PROMPT_TEMPLATE, type UpdateTestSuiteItem, type UpdateTestSuiteOptions, Usefulness, VercelAIChatModel, activateRunner, agentConfigContext, buildSuiteResult, createModel, createModelFromInstance, deserializeEvaluators, detectProvider, disableLogger, evaluate, evaluatePrompt, evaluateTestSuite, flushAll, generateId, getGlobalClient, getTrackContext, logger, resetGlobalClient, resolveEvaluators, resolveExecutionPolicy, resolveItemExecutionPolicy, resolveModel, runTests, serializeEvaluators, setGlobalClient, setLoggerLevel, track, validateEvaluators, validateExecutionPolicy };

package/dist/index.d.ts

@@ -12672,6 +12672,28 @@ type EvaluationTestResult = {
  * Mirrors the Fern-generated ExecutionPolicyWrite type.
  */
 type ExecutionPolicy = ExecutionPolicyWrite;
+/**
+ * A raw test suite item as returned by {@link TestSuite.getRawItems}.
+ *
+ * Unlike the view returned by {@link TestSuite.getItems}, this preserves:
+ * - `evaluators` as raw {@link EvaluatorItemWrite} objects (not decoded to assertion strings)
+ * - `executionPolicy` as the item-level value only (not merged with the suite-level default)
+ *
+ * `data` is the full stored payload exactly as returned by the dataset
+ * (so if an item was stored with a `description`, it will be present in
+ * `data` as well). `description` is additionally exposed as a top-level
+ * field for ergonomic access.
+ *
+ * Use this when you need to introspect or forward the stored evaluator
+ * config or per-item execution policy verbatim.
+ */
+interface RawTestSuiteItem {
+    id: string;
+    data: DatasetItemData;
+    description?: string;
+    evaluators?: EvaluatorItemWrite[];
+    executionPolicy?: ExecutionPolicy;
+}
 declare const DEFAULT_EXECUTION_POLICY: Required<ExecutionPolicy>;
 /**
  * A single item to be inserted into a test suite via `insert()`.
@@ -12854,6 +12876,24 @@ declare class TestSuite {
         assertions: string[];
         executionPolicy: Required<ExecutionPolicy>;
     }>>;
+    /**
+     * Retrieve items with full suite-specific metadata preserved verbatim.
+     *
+     * Unlike {@link getItems}, this does NOT:
+     * - Decode `evaluators` into assertion strings (you get raw {@link EvaluatorItemWrite}[])
+     * - Merge the item-level `executionPolicy` with the suite-level default
+     *
+     * `data` mirrors the stored payload (includes `description` if present);
+     * `description` is additionally exposed at the top level for convenience.
+     *
+     * Use this when you need to inspect or forward the stored evaluator
+     * config or per-item execution policy as-is.
+     *
+     * @param nbSamples Max items to retrieve. Omit to return all items.
+     * @param lastRetrievedId Opaque cursor for pagination (last `id` from a previous call).
+     * @returns Array of {@link RawTestSuiteItem} objects.
+     */
+    getRawItems(nbSamples?: number, lastRetrievedId?: string): Promise<RawTestSuiteItem[]>;
     getGlobalAssertions(): Promise<string[]>;
     getTags(): Promise<string[]>;
     getItemsCount(): Promise<number | undefined>;
@@ -13151,6 +13191,25 @@ declare class Experiment {
     getUrl(): Promise<string>;
 }
 
+interface TestSuiteExperimentData extends ExperimentData {
+    passRate?: number;
+    passedCount?: number;
+    totalCount?: number;
+    assertionScores?: AssertionScoreAveragePublic[];
+}
+/**
+ * Represents an experiment run against a test suite. Extends `Experiment`
+ * with the aggregate assertion statistics the backend populates only for
+ * evaluation-suite experiments (null/undefined for regular dataset experiments).
+ */
+declare class TestSuiteExperiment extends Experiment {
+    readonly passRate?: number;
+    readonly passedCount?: number;
+    readonly totalCount?: number;
+    readonly assertionScores?: AssertionScoreAveragePublic[];
+    constructor(data: TestSuiteExperimentData, opik: OpikClient);
+}
+
 interface ChatPromptData extends BasePromptData {
     messages: ChatMessage[];
 }
@@ -13596,6 +13655,25 @@ declare class OpikClient {
      * @throws {DatasetNotFoundError} If the dataset doesn't exist
      */
     getDatasetExperiments: (datasetName: string, maxResults?: number, projectName?: string) => Promise<Experiment[]>;
+    /**
+     * Retrieves all experiments associated with a test suite.
+     *
+     * @param name The name of the test suite
+     * @param maxResults Maximum number of experiments to return (default: 100)
+     * @param projectName Optional project name to scope the suite lookup. If not provided, uses the client's configured project.
+     * @returns A list of TestSuiteExperiment objects associated with the test suite,
+     *   each carrying the suite-specific assertion aggregates (`passRate`, `passedCount`,
+     *   `totalCount`, `assertionScores`) populated by the backend.
+     * @throws {DatasetNotFoundError} If the test suite doesn't exist
+     */
+    getTestSuiteExperiments: (name: string, maxResults?: number, projectName?: string) => Promise<TestSuiteExperiment[]>;
+    /**
+     * Paginated fetch of experiments for a given dataset ID, mapping each raw
+     * `ExperimentPublic` row to a caller-chosen entity. Used internally by
+     * `getDatasetExperiments` and `getTestSuiteExperiments` to share the
+     * loop shape and only differ on the constructed type.
+     */
+    private findExperimentsByDatasetId;
     /**
      * Deletes an experiment by ID
      *
@@ -16013,4 +16091,4 @@ declare class ConfigMismatchError extends OpikError {
 
 declare function activateRunner(): void;
 
-export { AgentTaskCompletionJudge, AgentToolCorrectnessJudge, type AllProviderOptions, AnnotationQueuePublicScope as AnnotationQueueScope, AnswerRelevance, type AnthropicProviderOptions, BaseLLMJudgeMetric, BaseMetric, BaseSuiteEvaluator, ChatPrompt, ComplianceRiskJudge, type Config, ConfigMismatchError, ConfigNotFoundError, Contains, type CreateTestSuiteOptions, DEFAULT_EXECUTION_POLICY, Dataset, type DatasetPublic, DatasetVersion, DatasetVersionNotFoundError, type DatasetVersionPublic, DemographicBiasJudge, DialogueHelpfulnessJudge, type ErrorInfo, type EvaluateOptions, type EvaluatePromptOptions, type EvaluateTestSuiteOptions, type EvaluationError, type EvaluationResult, type EvaluationScoreResult, type EvaluationTask, type EvaluationTestCase, type EvaluationTestResult, ExactMatch, type ExecutionPolicy, type FeedbackScoreData, type FewShotExampleAnswerRelevanceNoContext, type FewShotExampleAnswerRelevanceWithContext, type FewShotExampleHallucination, type FewShotExampleModeration, type FilterExpression, GEval, GEvalPreset, GenderBiasJudge, type GoogleProviderOptions, Hallucination, IsJson, type ItemResult, LLMJudge, type LLMJudgeConfig, type LLMJudgeModelSettings, type LLMJudgeOptions, type LLMJudgeResponseFormat, ModelConfigurationError, ModelError, ModelGenerationError, Moderation, type OpenAIProviderOptions, OpikClient as Opik, type OpikAssistantMessage, OpikBaseModel, type OpikConfig, type OpikMessage, OpikQueryLanguage, SpanType as OpikSpanType, type OpikSystemMessage, type OpikToolMessage, type OpikUserMessage, type Param, PoliticalBiasJudge, Prompt, PromptType, PromptUncertaintyJudge, type ProviderOptionsForModel, QARelevanceJudge, RegexMatch, RegionalBiasJudge, type RegistryEntry, ReligiousBiasJudge, ResponseSchema, type RunTestsOptions, SYSTEM_PROMPT, type ScoringKeyMappingType, Span, SpanType, SummarizationCoherenceJudge, SummarizationConsistencyJudge, type SupportedModelId, TestSuite, type TestSuiteItem, TestSuiteResult, ThreadsAnnotationQueue, Trace, TracesAnnotationQueue, USER_PROMPT_TEMPLATE, type UpdateTestSuiteItem, type UpdateTestSuiteOptions, Usefulness, VercelAIChatModel, activateRunner, agentConfigContext, buildSuiteResult, createModel, createModelFromInstance, deserializeEvaluators, detectProvider, disableLogger, evaluate, evaluatePrompt, evaluateTestSuite, flushAll, generateId, getGlobalClient, getTrackContext, logger, resetGlobalClient, resolveEvaluators, resolveExecutionPolicy, resolveItemExecutionPolicy, resolveModel, runTests, serializeEvaluators, setGlobalClient, setLoggerLevel, track, validateEvaluators, validateExecutionPolicy };
+export { AgentTaskCompletionJudge, AgentToolCorrectnessJudge, type AllProviderOptions, AnnotationQueuePublicScope as AnnotationQueueScope, AnswerRelevance, type AnthropicProviderOptions, BaseLLMJudgeMetric, BaseMetric, BaseSuiteEvaluator, ChatPrompt, ComplianceRiskJudge, type Config, ConfigMismatchError, ConfigNotFoundError, Contains, type CreateTestSuiteOptions, DEFAULT_EXECUTION_POLICY, Dataset, type DatasetPublic, DatasetVersion, DatasetVersionNotFoundError, type DatasetVersionPublic, DemographicBiasJudge, DialogueHelpfulnessJudge, type ErrorInfo, type EvaluateOptions, type EvaluatePromptOptions, type EvaluateTestSuiteOptions, type EvaluationError, type EvaluationResult, type EvaluationScoreResult, type EvaluationTask, type EvaluationTestCase, type EvaluationTestResult, ExactMatch, type ExecutionPolicy, type FeedbackScoreData, type FewShotExampleAnswerRelevanceNoContext, type FewShotExampleAnswerRelevanceWithContext, type FewShotExampleHallucination, type FewShotExampleModeration, type FilterExpression, GEval, GEvalPreset, GenderBiasJudge, type GoogleProviderOptions, Hallucination, IsJson, type ItemResult, LLMJudge, type LLMJudgeConfig, type LLMJudgeModelSettings, type LLMJudgeOptions, type LLMJudgeResponseFormat, ModelConfigurationError, ModelError, ModelGenerationError, Moderation, type OpenAIProviderOptions, OpikClient as Opik, type OpikAssistantMessage, OpikBaseModel, type OpikConfig, type OpikMessage, OpikQueryLanguage, SpanType as OpikSpanType, type OpikSystemMessage, type OpikToolMessage, type OpikUserMessage, type Param, PoliticalBiasJudge, Prompt, PromptType, PromptUncertaintyJudge, type ProviderOptionsForModel, QARelevanceJudge, type RawTestSuiteItem, RegexMatch, RegionalBiasJudge, type RegistryEntry, ReligiousBiasJudge, ResponseSchema, type RunTestsOptions, SYSTEM_PROMPT, type ScoringKeyMappingType, Span, SpanType, SummarizationCoherenceJudge, SummarizationConsistencyJudge, type SupportedModelId, TestSuite, type TestSuiteItem, TestSuiteResult, ThreadsAnnotationQueue, Trace, TracesAnnotationQueue, USER_PROMPT_TEMPLATE, type UpdateTestSuiteItem, type UpdateTestSuiteOptions, Usefulness, VercelAIChatModel, activateRunner, agentConfigContext, buildSuiteResult, createModel, createModelFromInstance, deserializeEvaluators, detectProvider, disableLogger, evaluate, evaluatePrompt, evaluateTestSuite, flushAll, generateId, getGlobalClient, getTrackContext, logger, resetGlobalClient, resolveEvaluators, resolveExecutionPolicy, resolveItemExecutionPolicy, resolveModel, runTests, serializeEvaluators, setGlobalClient, setLoggerLevel, track, validateEvaluators, validateExecutionPolicy };

package/dist/index.js

@@ -1 +1 @@
-import {Ba}from'./chunk-FTTQJ4TO.js';export{ra as AgentTaskCompletionJudge,qa as AgentToolCorrectnessJudge,ea as AnswerRelevance,aa as BaseLLMJudgeMetric,B as BaseMetric,C as BaseSuiteEvaluator,p as ChatPrompt,ta as ComplianceRiskJudge,i as ConfigMismatchError,h as ConfigNotFoundError,Z as Contains,y as DEFAULT_EXECUTION_POLICY,j as Dataset,f as DatasetVersion,g as DatasetVersionNotFoundError,la as DemographicBiasJudge,ja as DialogueHelpfulnessJudge,Y as ExactMatch,fa as GEval,ga as GEvalPreset,na as GenderBiasJudge,da as Hallucination,$ as IsJson,P as LLMJudge,G as ModelConfigurationError,E as ModelError,F as ModelGenerationError,ba as Moderation,za as Opik,D as OpikBaseModel,q as OpikQueryLanguage,d as OpikSpanType,ma as PoliticalBiasJudge,o as Prompt,k as PromptType,sa as PromptUncertaintyJudge,ka as QARelevanceJudge,_ as RegexMatch,pa as RegionalBiasJudge,oa as ReligiousBiasJudge,O as ResponseSchema,M as SYSTEM_PROMPT,ia as SummarizationCoherenceJudge,ha as SummarizationConsistencyJudge,xa as TestSuite,z as TestSuiteResult,s as ThreadsAnnotationQueue,r as TracesAnnotationQueue,N as USER_PROMPT_TEMPLATE,ca as Usefulness,I as VercelAIChatModel,v as activateRunner,t as agentConfigContext,A as buildSuiteResult,J as createModel,K as createModelFromInstance,R as deserializeEvaluators,H as detectProvider,c as disableLogger,W as evaluate,X as evaluatePrompt,U as evaluateTestSuite,u as flushAll,e as generateId,l as getGlobalClient,w as getTrackContext,a as logger,n as resetGlobalClient,ua as resolveEvaluators,S as resolveExecutionPolicy,T as resolveItemExecutionPolicy,L as resolveModel,V as runTests,Q as serializeEvaluators,m as setGlobalClient,b as setLoggerLevel,x as track,va as validateEvaluators,wa as validateExecutionPolicy,Aa as z}from'./chunk-FTTQJ4TO.js';Ba();
+import {Ba}from'./chunk-GRBNDL4U.js';export{ra as AgentTaskCompletionJudge,qa as AgentToolCorrectnessJudge,ea as AnswerRelevance,aa as BaseLLMJudgeMetric,B as BaseMetric,C as BaseSuiteEvaluator,p as ChatPrompt,ta as ComplianceRiskJudge,i as ConfigMismatchError,h as ConfigNotFoundError,Z as Contains,y as DEFAULT_EXECUTION_POLICY,j as Dataset,f as DatasetVersion,g as DatasetVersionNotFoundError,la as DemographicBiasJudge,ja as DialogueHelpfulnessJudge,Y as ExactMatch,fa as GEval,ga as GEvalPreset,na as GenderBiasJudge,da as Hallucination,$ as IsJson,P as LLMJudge,G as ModelConfigurationError,E as ModelError,F as ModelGenerationError,ba as Moderation,za as Opik,D as OpikBaseModel,q as OpikQueryLanguage,d as OpikSpanType,ma as PoliticalBiasJudge,o as Prompt,k as PromptType,sa as PromptUncertaintyJudge,ka as QARelevanceJudge,_ as RegexMatch,pa as RegionalBiasJudge,oa as ReligiousBiasJudge,O as ResponseSchema,M as SYSTEM_PROMPT,ia as SummarizationCoherenceJudge,ha as SummarizationConsistencyJudge,xa as TestSuite,z as TestSuiteResult,s as ThreadsAnnotationQueue,r as TracesAnnotationQueue,N as USER_PROMPT_TEMPLATE,ca as Usefulness,I as VercelAIChatModel,v as activateRunner,t as agentConfigContext,A as buildSuiteResult,J as createModel,K as createModelFromInstance,R as deserializeEvaluators,H as detectProvider,c as disableLogger,W as evaluate,X as evaluatePrompt,U as evaluateTestSuite,u as flushAll,e as generateId,l as getGlobalClient,w as getTrackContext,a as logger,n as resetGlobalClient,ua as resolveEvaluators,S as resolveExecutionPolicy,T as resolveItemExecutionPolicy,L as resolveModel,V as runTests,Q as serializeEvaluators,m as setGlobalClient,b as setLoggerLevel,x as track,va as validateEvaluators,wa as validateExecutionPolicy,Aa as z}from'./chunk-GRBNDL4U.js';Ba();

package/dist/{suite-2DEGP66I.js → suite-QBCPCDGQ.js}

@@ -1 +1 @@
-import {ya}from'./chunk-FTTQJ4TO.js';export{y as DEFAULT_EXECUTION_POLICY,xa as TestSuite,z as TestSuiteResult,A as buildSuiteResult,R as deserializeEvaluators,U as evaluateTestSuite,S as resolveExecutionPolicy,T as resolveItemExecutionPolicy,V as runTests,Q as serializeEvaluators}from'./chunk-FTTQJ4TO.js';ya();
+import {ya}from'./chunk-GRBNDL4U.js';export{y as DEFAULT_EXECUTION_POLICY,xa as TestSuite,z as TestSuiteResult,A as buildSuiteResult,R as deserializeEvaluators,U as evaluateTestSuite,S as resolveExecutionPolicy,T as resolveItemExecutionPolicy,V as runTests,Q as serializeEvaluators}from'./chunk-GRBNDL4U.js';ya();

package/package.json

@@ -1,7 +1,7 @@
 {
   "name": "opik",
   "description": "Opik TypeScript and JavaScript SDK",
-  "version": "2.0.7",
+  "version": "2.0.9",
   "repository": {
     "type": "git",
     "url": "git+https://github.com/comet-ml/opik.git",