openai 4.97.0 → 4.99.0

package/resources/evals/evals.d.ts

@@ -1,6 +1,7 @@
 import { APIResource } from "../../resource.js";
 import * as Core from "../../core.js";
 import * as Shared from "../shared.js";
+import * as GraderModelsAPI from "../graders/grader-models.js";
 import * as ResponsesAPI from "../responses/responses.js";
 import * as RunsAPI from "./runs/runs.js";
 import { CreateEvalCompletionsRunDataSource, CreateEvalJSONLRunDataSource, EvalAPIError, RunCancelResponse, RunCreateParams, RunCreateResponse, RunDeleteResponse, RunListParams, RunListResponse, RunListResponsesPage, RunRetrieveResponse, Runs } from "./runs/runs.js";
@@ -55,77 +56,7 @@ export interface EvalCustomDataSourceConfig {
     type: 'custom';
 }
 /**
- * A LabelModelGrader object which uses a model to assign labels to each item in
- * the evaluation.
- */
-export interface EvalLabelModelGrader {
-    input: Array<EvalLabelModelGrader.Input>;
-    /**
-     * The labels to assign to each item in the evaluation.
-     */
-    labels: Array<string>;
-    /**
-     * The model to use for the evaluation. Must support structured outputs.
-     */
-    model: string;
-    /**
-     * The name of the grader.
-     */
-    name: string;
-    /**
-     * The labels that indicate a passing result. Must be a subset of labels.
-     */
-    passing_labels: Array<string>;
-    /**
-     * The object type, which is always `label_model`.
-     */
-    type: 'label_model';
-}
-export declare namespace EvalLabelModelGrader {
-    /**
-     * A message input to the model with a role indicating instruction following
-     * hierarchy. Instructions given with the `developer` or `system` role take
-     * precedence over instructions given with the `user` role. Messages with the
-     * `assistant` role are presumed to have been generated by the model in previous
-     * interactions.
-     */
-    interface Input {
-        /**
-         * Text inputs to the model - can contain template strings.
-         */
-        content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-        /**
-         * The role of the message input. One of `user`, `assistant`, `system`, or
-         * `developer`.
-         */
-        role: 'user' | 'assistant' | 'system' | 'developer';
-        /**
-         * The type of the message input. Always `message`.
-         */
-        type?: 'message';
-    }
-    namespace Input {
-        /**
-         * A text output from the model.
-         */
-        interface OutputText {
-            /**
-             * The text output from the model.
-             */
-            text: string;
-            /**
-             * The type of the output text. Always `output_text`.
-             */
-            type: 'output_text';
-        }
-    }
-}
-/**
- * A StoredCompletionsDataSourceConfig which specifies the metadata property of
- * your stored completions query. This is usually metadata like `usecase=chatbot`
- * or `prompt-version=v2`, etc. The schema returned by this data source config is
- * used to defined what variables are available in your evals. `item` and `sample`
- * are both defined when using this data source config.
+ * @deprecated Deprecated in favor of LogsDataSourceConfig.
  */
 export interface EvalStoredCompletionsDataSourceConfig {
     /**
@@ -134,9 +65,9 @@ export interface EvalStoredCompletionsDataSourceConfig {
      */
     schema: Record<string, unknown>;
     /**
-     * The type of data source. Always `stored_completions`.
+     * The type of data source. Always `stored-completions`.
      */
-    type: 'stored_completions';
+    type: 'stored-completions';
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -147,62 +78,6 @@ export interface EvalStoredCompletionsDataSourceConfig {
      */
     metadata?: Shared.Metadata | null;
 }
-/**
- * A StringCheckGrader object that performs a string comparison between input and
- * reference using a specified operation.
- */
-export interface EvalStringCheckGrader {
-    /**
-     * The input text. This may include template strings.
-     */
-    input: string;
-    /**
-     * The name of the grader.
-     */
-    name: string;
-    /**
-     * The string check operation to perform. One of `eq`, `ne`, `like`, or `ilike`.
-     */
-    operation: 'eq' | 'ne' | 'like' | 'ilike';
-    /**
-     * The reference text. This may include template strings.
-     */
-    reference: string;
-    /**
-     * The object type, which is always `string_check`.
-     */
-    type: 'string_check';
-}
-/**
- * A TextSimilarityGrader object which grades text based on similarity metrics.
- */
-export interface EvalTextSimilarityGrader {
-    /**
-     * The evaluation metric to use. One of `fuzzy_match`, `bleu`, `gleu`, `meteor`,
-     * `rouge_1`, `rouge_2`, `rouge_3`, `rouge_4`, `rouge_5`, or `rouge_l`.
-     */
-    evaluation_metric: 'fuzzy_match' | 'bleu' | 'gleu' | 'meteor' | 'rouge_1' | 'rouge_2' | 'rouge_3' | 'rouge_4' | 'rouge_5' | 'rouge_l';
-    /**
-     * The text being graded.
-     */
-    input: string;
-    /**
-     * A float score where a value greater than or equal indicates a passing grade.
-     */
-    pass_threshold: number;
-    /**
-     * The text being graded against.
-     */
-    reference: string;
-    /**
-     * The type of grader.
-     */
-    type: 'text_similarity';
-    /**
-     * The name of the grader.
-     */
-    name?: string;
-}
 /**
  * An Eval object with a data source config and testing criteria. An Eval
  * represents a task to be done for your LLM integration. Like:
@@ -223,7 +98,7 @@ export interface EvalCreateResponse {
     /**
      * Configuration of data sources used in runs of the evaluation.
      */
-    data_source_config: EvalCustomDataSourceConfig | EvalStoredCompletionsDataSourceConfig;
+    data_source_config: EvalCustomDataSourceConfig | EvalCreateResponse.Logs | EvalStoredCompletionsDataSourceConfig;
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -244,29 +119,49 @@ export interface EvalCreateResponse {
     /**
      * A list of testing criteria.
      */
-    testing_criteria: Array<EvalLabelModelGrader | EvalStringCheckGrader | EvalTextSimilarityGrader | EvalCreateResponse.Python | EvalCreateResponse.ScoreModel>;
+    testing_criteria: Array<GraderModelsAPI.LabelModelGrader | GraderModelsAPI.StringCheckGrader | EvalCreateResponse.EvalGraderTextSimilarity | EvalCreateResponse.EvalGraderPython | EvalCreateResponse.EvalGraderScoreModel>;
 }
 export declare namespace EvalCreateResponse {
     /**
-     * A PythonGrader object that runs a python script on the input.
+     * A LogsDataSourceConfig which specifies the metadata property of your logs query.
+     * This is usually metadata like `usecase=chatbot` or `prompt-version=v2`, etc. The
+     * schema returned by this data source config is used to defined what variables are
+     * available in your evals. `item` and `sample` are both defined when using this
+     * data source config.
      */
-    interface Python {
+    interface Logs {
         /**
-         * The name of the grader.
+         * The json schema for the run data source items. Learn how to build JSON schemas
+         * [here](https://json-schema.org/).
          */
-        name: string;
+        schema: Record<string, unknown>;
         /**
-         * The source code of the python script.
+         * The type of data source. Always `logs`.
          */
-        source: string;
+        type: 'logs';
         /**
-         * The object type, which is always `python`.
+         * Set of 16 key-value pairs that can be attached to an object. This can be useful
+         * for storing additional information about the object in a structured format, and
+         * querying for objects via API or the dashboard.
+         *
+         * Keys are strings with a maximum length of 64 characters. Values are strings with
+         * a maximum length of 512 characters.
          */
-        type: 'python';
+        metadata?: Shared.Metadata | null;
+    }
+    /**
+     * A TextSimilarityGrader object which grades text based on similarity metrics.
+     */
+    interface EvalGraderTextSimilarity extends GraderModelsAPI.TextSimilarityGrader {
         /**
-         * The image tag to use for the python script.
+         * The threshold for the score.
          */
-        image_tag?: string;
+        pass_threshold: number;
+    }
+    /**
+     * A PythonGrader object that runs a python script on the input.
+     */
+    interface EvalGraderPython extends GraderModelsAPI.PythonGrader {
         /**
          * The threshold for the score.
          */
@@ -275,74 +170,11 @@ export declare namespace EvalCreateResponse {
     /**
      * A ScoreModelGrader object that uses a model to assign a score to the input.
      */
-    interface ScoreModel {
-        /**
-         * The input text. This may include template strings.
-         */
-        input: Array<ScoreModel.Input>;
-        /**
-         * The model to use for the evaluation.
-         */
-        model: string;
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The object type, which is always `score_model`.
-         */
-        type: 'score_model';
+    interface EvalGraderScoreModel extends GraderModelsAPI.ScoreModelGrader {
         /**
          * The threshold for the score.
          */
         pass_threshold?: number;
-        /**
-         * The range of the score. Defaults to `[0, 1]`.
-         */
-        range?: Array<number>;
-        /**
-         * The sampling parameters for the model.
-         */
-        sampling_params?: unknown;
-    }
-    namespace ScoreModel {
-        /**
-         * A message input to the model with a role indicating instruction following
-         * hierarchy. Instructions given with the `developer` or `system` role take
-         * precedence over instructions given with the `user` role. Messages with the
-         * `assistant` role are presumed to have been generated by the model in previous
-         * interactions.
-         */
-        interface Input {
-            /**
-             * Text inputs to the model - can contain template strings.
-             */
-            content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-            /**
-             * The role of the message input. One of `user`, `assistant`, `system`, or
-             * `developer`.
-             */
-            role: 'user' | 'assistant' | 'system' | 'developer';
-            /**
-             * The type of the message input. Always `message`.
-             */
-            type?: 'message';
-        }
-        namespace Input {
-            /**
-             * A text output from the model.
-             */
-            interface OutputText {
-                /**
-                 * The text output from the model.
-                 */
-                text: string;
-                /**
-                 * The type of the output text. Always `output_text`.
-                 */
-                type: 'output_text';
-            }
-        }
     }
 }
 /**
@@ -365,7 +197,7 @@ export interface EvalRetrieveResponse {
     /**
      * Configuration of data sources used in runs of the evaluation.
      */
-    data_source_config: EvalCustomDataSourceConfig | EvalStoredCompletionsDataSourceConfig;
+    data_source_config: EvalCustomDataSourceConfig | EvalRetrieveResponse.Logs | EvalStoredCompletionsDataSourceConfig;
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -386,29 +218,49 @@ export interface EvalRetrieveResponse {
     /**
      * A list of testing criteria.
      */
-    testing_criteria: Array<EvalLabelModelGrader | EvalStringCheckGrader | EvalTextSimilarityGrader | EvalRetrieveResponse.Python | EvalRetrieveResponse.ScoreModel>;
+    testing_criteria: Array<GraderModelsAPI.LabelModelGrader | GraderModelsAPI.StringCheckGrader | EvalRetrieveResponse.EvalGraderTextSimilarity | EvalRetrieveResponse.EvalGraderPython | EvalRetrieveResponse.EvalGraderScoreModel>;
 }
 export declare namespace EvalRetrieveResponse {
     /**
-     * A PythonGrader object that runs a python script on the input.
+     * A LogsDataSourceConfig which specifies the metadata property of your logs query.
+     * This is usually metadata like `usecase=chatbot` or `prompt-version=v2`, etc. The
+     * schema returned by this data source config is used to defined what variables are
+     * available in your evals. `item` and `sample` are both defined when using this
+     * data source config.
      */
-    interface Python {
+    interface Logs {
         /**
-         * The name of the grader.
+         * The json schema for the run data source items. Learn how to build JSON schemas
+         * [here](https://json-schema.org/).
          */
-        name: string;
+        schema: Record<string, unknown>;
         /**
-         * The source code of the python script.
+         * The type of data source. Always `logs`.
          */
-        source: string;
+        type: 'logs';
         /**
-         * The object type, which is always `python`.
+         * Set of 16 key-value pairs that can be attached to an object. This can be useful
+         * for storing additional information about the object in a structured format, and
+         * querying for objects via API or the dashboard.
+         *
+         * Keys are strings with a maximum length of 64 characters. Values are strings with
+         * a maximum length of 512 characters.
          */
-        type: 'python';
+        metadata?: Shared.Metadata | null;
+    }
+    /**
+     * A TextSimilarityGrader object which grades text based on similarity metrics.
+     */
+    interface EvalGraderTextSimilarity extends GraderModelsAPI.TextSimilarityGrader {
         /**
-         * The image tag to use for the python script.
+         * The threshold for the score.
          */
-        image_tag?: string;
+        pass_threshold: number;
+    }
+    /**
+     * A PythonGrader object that runs a python script on the input.
+     */
+    interface EvalGraderPython extends GraderModelsAPI.PythonGrader {
         /**
          * The threshold for the score.
          */
@@ -417,74 +269,11 @@ export declare namespace EvalRetrieveResponse {
     /**
      * A ScoreModelGrader object that uses a model to assign a score to the input.
      */
-    interface ScoreModel {
-        /**
-         * The input text. This may include template strings.
-         */
-        input: Array<ScoreModel.Input>;
-        /**
-         * The model to use for the evaluation.
-         */
-        model: string;
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The object type, which is always `score_model`.
-         */
-        type: 'score_model';
+    interface EvalGraderScoreModel extends GraderModelsAPI.ScoreModelGrader {
         /**
          * The threshold for the score.
          */
         pass_threshold?: number;
-        /**
-         * The range of the score. Defaults to `[0, 1]`.
-         */
-        range?: Array<number>;
-        /**
-         * The sampling parameters for the model.
-         */
-        sampling_params?: unknown;
-    }
-    namespace ScoreModel {
-        /**
-         * A message input to the model with a role indicating instruction following
-         * hierarchy. Instructions given with the `developer` or `system` role take
-         * precedence over instructions given with the `user` role. Messages with the
-         * `assistant` role are presumed to have been generated by the model in previous
-         * interactions.
-         */
-        interface Input {
-            /**
-             * Text inputs to the model - can contain template strings.
-             */
-            content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-            /**
-             * The role of the message input. One of `user`, `assistant`, `system`, or
-             * `developer`.
-             */
-            role: 'user' | 'assistant' | 'system' | 'developer';
-            /**
-             * The type of the message input. Always `message`.
-             */
-            type?: 'message';
-        }
-        namespace Input {
-            /**
-             * A text output from the model.
-             */
-            interface OutputText {
-                /**
-                 * The text output from the model.
-                 */
-                text: string;
-                /**
-                 * The type of the output text. Always `output_text`.
-                 */
-                type: 'output_text';
-            }
-        }
     }
 }
 /**
@@ -507,7 +296,7 @@ export interface EvalUpdateResponse {
     /**
      * Configuration of data sources used in runs of the evaluation.
      */
-    data_source_config: EvalCustomDataSourceConfig | EvalStoredCompletionsDataSourceConfig;
+    data_source_config: EvalCustomDataSourceConfig | EvalUpdateResponse.Logs | EvalStoredCompletionsDataSourceConfig;
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -528,29 +317,49 @@ export interface EvalUpdateResponse {
     /**
      * A list of testing criteria.
      */
-    testing_criteria: Array<EvalLabelModelGrader | EvalStringCheckGrader | EvalTextSimilarityGrader | EvalUpdateResponse.Python | EvalUpdateResponse.ScoreModel>;
+    testing_criteria: Array<GraderModelsAPI.LabelModelGrader | GraderModelsAPI.StringCheckGrader | EvalUpdateResponse.EvalGraderTextSimilarity | EvalUpdateResponse.EvalGraderPython | EvalUpdateResponse.EvalGraderScoreModel>;
 }
 export declare namespace EvalUpdateResponse {
     /**
-     * A PythonGrader object that runs a python script on the input.
+     * A LogsDataSourceConfig which specifies the metadata property of your logs query.
+     * This is usually metadata like `usecase=chatbot` or `prompt-version=v2`, etc. The
+     * schema returned by this data source config is used to defined what variables are
+     * available in your evals. `item` and `sample` are both defined when using this
+     * data source config.
      */
-    interface Python {
+    interface Logs {
         /**
-         * The name of the grader.
+         * The json schema for the run data source items. Learn how to build JSON schemas
+         * [here](https://json-schema.org/).
          */
-        name: string;
+        schema: Record<string, unknown>;
         /**
-         * The source code of the python script.
+         * The type of data source. Always `logs`.
          */
-        source: string;
+        type: 'logs';
         /**
-         * The object type, which is always `python`.
+         * Set of 16 key-value pairs that can be attached to an object. This can be useful
+         * for storing additional information about the object in a structured format, and
+         * querying for objects via API or the dashboard.
+         *
+         * Keys are strings with a maximum length of 64 characters. Values are strings with
+         * a maximum length of 512 characters.
          */
-        type: 'python';
+        metadata?: Shared.Metadata | null;
+    }
+    /**
+     * A TextSimilarityGrader object which grades text based on similarity metrics.
+     */
+    interface EvalGraderTextSimilarity extends GraderModelsAPI.TextSimilarityGrader {
         /**
-         * The image tag to use for the python script.
+         * The threshold for the score.
          */
-        image_tag?: string;
+        pass_threshold: number;
+    }
+    /**
+     * A PythonGrader object that runs a python script on the input.
+     */
+    interface EvalGraderPython extends GraderModelsAPI.PythonGrader {
         /**
          * The threshold for the score.
          */
@@ -559,74 +368,11 @@ export declare namespace EvalUpdateResponse {
     /**
      * A ScoreModelGrader object that uses a model to assign a score to the input.
      */
-    interface ScoreModel {
-        /**
-         * The input text. This may include template strings.
-         */
-        input: Array<ScoreModel.Input>;
-        /**
-         * The model to use for the evaluation.
-         */
-        model: string;
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The object type, which is always `score_model`.
-         */
-        type: 'score_model';
+    interface EvalGraderScoreModel extends GraderModelsAPI.ScoreModelGrader {
         /**
          * The threshold for the score.
          */
         pass_threshold?: number;
-        /**
-         * The range of the score. Defaults to `[0, 1]`.
-         */
-        range?: Array<number>;
-        /**
-         * The sampling parameters for the model.
-         */
-        sampling_params?: unknown;
-    }
-    namespace ScoreModel {
-        /**
-         * A message input to the model with a role indicating instruction following
-         * hierarchy. Instructions given with the `developer` or `system` role take
-         * precedence over instructions given with the `user` role. Messages with the
-         * `assistant` role are presumed to have been generated by the model in previous
-         * interactions.
-         */
-        interface Input {
-            /**
-             * Text inputs to the model - can contain template strings.
-             */
-            content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-            /**
-             * The role of the message input. One of `user`, `assistant`, `system`, or
-             * `developer`.
-             */
-            role: 'user' | 'assistant' | 'system' | 'developer';
-            /**
-             * The type of the message input. Always `message`.
-             */
-            type?: 'message';
-        }
-        namespace Input {
-            /**
-             * A text output from the model.
-             */
-            interface OutputText {
-                /**
-                 * The text output from the model.
-                 */
-                text: string;
-                /**
-                 * The type of the output text. Always `output_text`.
-                 */
-                type: 'output_text';
-            }
-        }
     }
 }
 /**
@@ -649,7 +395,7 @@ export interface EvalListResponse {
     /**
      * Configuration of data sources used in runs of the evaluation.
      */
-    data_source_config: EvalCustomDataSourceConfig | EvalStoredCompletionsDataSourceConfig;
+    data_source_config: EvalCustomDataSourceConfig | EvalListResponse.Logs | EvalStoredCompletionsDataSourceConfig;
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -670,29 +416,49 @@ export interface EvalListResponse {
     /**
      * A list of testing criteria.
      */
-    testing_criteria: Array<EvalLabelModelGrader | EvalStringCheckGrader | EvalTextSimilarityGrader | EvalListResponse.Python | EvalListResponse.ScoreModel>;
+    testing_criteria: Array<GraderModelsAPI.LabelModelGrader | GraderModelsAPI.StringCheckGrader | EvalListResponse.EvalGraderTextSimilarity | EvalListResponse.EvalGraderPython | EvalListResponse.EvalGraderScoreModel>;
 }
 export declare namespace EvalListResponse {
     /**
-     * A PythonGrader object that runs a python script on the input.
+     * A LogsDataSourceConfig which specifies the metadata property of your logs query.
+     * This is usually metadata like `usecase=chatbot` or `prompt-version=v2`, etc. The
+     * schema returned by this data source config is used to defined what variables are
+     * available in your evals. `item` and `sample` are both defined when using this
+     * data source config.
      */
-    interface Python {
+    interface Logs {
         /**
-         * The name of the grader.
+         * The json schema for the run data source items. Learn how to build JSON schemas
+         * [here](https://json-schema.org/).
          */
-        name: string;
+        schema: Record<string, unknown>;
         /**
-         * The source code of the python script.
+         * The type of data source. Always `logs`.
          */
-        source: string;
+        type: 'logs';
         /**
-         * The object type, which is always `python`.
+         * Set of 16 key-value pairs that can be attached to an object. This can be useful
+         * for storing additional information about the object in a structured format, and
+         * querying for objects via API or the dashboard.
+         *
+         * Keys are strings with a maximum length of 64 characters. Values are strings with
+         * a maximum length of 512 characters.
          */
-        type: 'python';
+        metadata?: Shared.Metadata | null;
+    }
+    /**
+     * A TextSimilarityGrader object which grades text based on similarity metrics.
+     */
+    interface EvalGraderTextSimilarity extends GraderModelsAPI.TextSimilarityGrader {
         /**
-         * The image tag to use for the python script.
+         * The threshold for the score.
          */
-        image_tag?: string;
+        pass_threshold: number;
+    }
+    /**
+     * A PythonGrader object that runs a python script on the input.
+     */
+    interface EvalGraderPython extends GraderModelsAPI.PythonGrader {
         /**
          * The threshold for the score.
          */
@@ -701,74 +467,11 @@ export declare namespace EvalListResponse {
     /**
      * A ScoreModelGrader object that uses a model to assign a score to the input.
      */
-    interface ScoreModel {
-        /**
-         * The input text. This may include template strings.
-         */
-        input: Array<ScoreModel.Input>;
-        /**
-         * The model to use for the evaluation.
-         */
-        model: string;
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The object type, which is always `score_model`.
-         */
-        type: 'score_model';
+    interface EvalGraderScoreModel extends GraderModelsAPI.ScoreModelGrader {
         /**
          * The threshold for the score.
          */
         pass_threshold?: number;
-        /**
-         * The range of the score. Defaults to `[0, 1]`.
-         */
-        range?: Array<number>;
-        /**
-         * The sampling parameters for the model.
-         */
-        sampling_params?: unknown;
-    }
-    namespace ScoreModel {
-        /**
-         * A message input to the model with a role indicating instruction following
-         * hierarchy. Instructions given with the `developer` or `system` role take
-         * precedence over instructions given with the `user` role. Messages with the
-         * `assistant` role are presumed to have been generated by the model in previous
-         * interactions.
-         */
-        interface Input {
-            /**
-             * Text inputs to the model - can contain template strings.
-             */
-            content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-            /**
-             * The role of the message input. One of `user`, `assistant`, `system`, or
-             * `developer`.
-             */
-            role: 'user' | 'assistant' | 'system' | 'developer';
-            /**
-             * The type of the message input. Always `message`.
-             */
-            type?: 'message';
-        }
-        namespace Input {
-            /**
-             * A text output from the model.
-             */
-            interface OutputText {
-                /**
-                 * The text output from the model.
-                 */
-                text: string;
-                /**
-                 * The type of the output text. Always `output_text`.
-                 */
-                type: 'output_text';
-            }
-        }
     }
 }
 export interface EvalDeleteResponse {
@@ -780,11 +483,11 @@ export interface EvalCreateParams {
     /**
      * The configuration for the data source used for the evaluation runs.
      */
-    data_source_config: EvalCreateParams.Custom | EvalCreateParams.Logs;
+    data_source_config: EvalCreateParams.Custom | EvalCreateParams.Logs | EvalCreateParams.StoredCompletions;
     /**
      * A list of graders for all eval runs in this group.
      */
-    testing_criteria: Array<EvalCreateParams.LabelModel | EvalStringCheckGrader | EvalTextSimilarityGrader | EvalCreateParams.Python | EvalCreateParams.ScoreModel>;
+    testing_criteria: Array<EvalCreateParams.LabelModel | GraderModelsAPI.StringCheckGrader | EvalCreateParams.TextSimilarity | EvalCreateParams.Python | EvalCreateParams.ScoreModel>;
     /**
      * Set of 16 key-value pairs that can be attached to an object. This can be useful
      * for storing additional information about the object in a structured format, and
@@ -824,9 +527,8 @@ export declare namespace EvalCreateParams {
         include_sample_schema?: boolean;
     }
     /**
-     * A data source config which specifies the metadata property of your stored
-     * completions query. This is usually metadata like `usecase=chatbot` or
-     * `prompt-version=v2`, etc.
+     * A data source config which specifies the metadata property of your logs query.
+     * This is usually metadata like `usecase=chatbot` or `prompt-version=v2`, etc.
      */
     interface Logs {
         /**
@@ -838,6 +540,19 @@ export declare namespace EvalCreateParams {
          */
         metadata?: Record<string, unknown>;
     }
+    /**
+     * Deprecated in favor of LogsDataSourceConfig.
+     */
+    interface StoredCompletions {
+        /**
+         * The type of data source. Always `stored-completions`.
+         */
+        type: 'stored-completions';
+        /**
+         * Metadata filters for the stored completions data source.
+         */
+        metadata?: Record<string, unknown>;
+    }
     /**
      * A LabelModelGrader object which uses a model to assign labels to each item in
      * the evaluation.
@@ -919,25 +634,18 @@ export declare namespace EvalCreateParams {
         }
     }
     /**
-     * A PythonGrader object that runs a python script on the input.
+     * A TextSimilarityGrader object which grades text based on similarity metrics.
      */
-    interface Python {
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The source code of the python script.
-         */
-        source: string;
+    interface TextSimilarity extends GraderModelsAPI.TextSimilarityGrader {
         /**
-         * The object type, which is always `python`.
-         */
-        type: 'python';
-        /**
-         * The image tag to use for the python script.
+         * The threshold for the score.
          */
-        image_tag?: string;
+        pass_threshold: number;
+    }
+    /**
+     * A PythonGrader object that runs a python script on the input.
+     */
+    interface Python extends GraderModelsAPI.PythonGrader {
         /**
          * The threshold for the score.
          */
@@ -946,74 +654,11 @@ export declare namespace EvalCreateParams {
     /**
      * A ScoreModelGrader object that uses a model to assign a score to the input.
      */
-    interface ScoreModel {
-        /**
-         * The input text. This may include template strings.
-         */
-        input: Array<ScoreModel.Input>;
-        /**
-         * The model to use for the evaluation.
-         */
-        model: string;
-        /**
-         * The name of the grader.
-         */
-        name: string;
-        /**
-         * The object type, which is always `score_model`.
-         */
-        type: 'score_model';
+    interface ScoreModel extends GraderModelsAPI.ScoreModelGrader {
         /**
          * The threshold for the score.
          */
         pass_threshold?: number;
-        /**
-         * The range of the score. Defaults to `[0, 1]`.
-         */
-        range?: Array<number>;
-        /**
-         * The sampling parameters for the model.
-         */
-        sampling_params?: unknown;
-    }
-    namespace ScoreModel {
-        /**
-         * A message input to the model with a role indicating instruction following
-         * hierarchy. Instructions given with the `developer` or `system` role take
-         * precedence over instructions given with the `user` role. Messages with the
-         * `assistant` role are presumed to have been generated by the model in previous
-         * interactions.
-         */
-        interface Input {
-            /**
-             * Text inputs to the model - can contain template strings.
-             */
-            content: string | ResponsesAPI.ResponseInputText | Input.OutputText;
-            /**
-             * The role of the message input. One of `user`, `assistant`, `system`, or
-             * `developer`.
-             */
-            role: 'user' | 'assistant' | 'system' | 'developer';
-            /**
-             * The type of the message input. Always `message`.
-             */
-            type?: 'message';
-        }
-        namespace Input {
-            /**
-             * A text output from the model.
-             */
-            interface OutputText {
-                /**
-                 * The text output from the model.
-                 */
-                text: string;
-                /**
-                 * The type of the output text. Always `output_text`.
-                 */
-                type: 'output_text';
-            }
-        }
     }
 }
 export interface EvalUpdateParams {
@@ -1044,7 +689,7 @@ export interface EvalListParams extends CursorPageParams {
     order_by?: 'created_at' | 'updated_at';
 }
 export declare namespace Evals {
-    export { type EvalCustomDataSourceConfig as EvalCustomDataSourceConfig, type EvalLabelModelGrader as EvalLabelModelGrader, type EvalStoredCompletionsDataSourceConfig as EvalStoredCompletionsDataSourceConfig, type EvalStringCheckGrader as EvalStringCheckGrader, type EvalTextSimilarityGrader as EvalTextSimilarityGrader, type EvalCreateResponse as EvalCreateResponse, type EvalRetrieveResponse as EvalRetrieveResponse, type EvalUpdateResponse as EvalUpdateResponse, type EvalListResponse as EvalListResponse, type EvalDeleteResponse as EvalDeleteResponse, EvalListResponsesPage as EvalListResponsesPage, type EvalCreateParams as EvalCreateParams, type EvalUpdateParams as EvalUpdateParams, type EvalListParams as EvalListParams, };
+    export { type EvalCustomDataSourceConfig as EvalCustomDataSourceConfig, type EvalStoredCompletionsDataSourceConfig as EvalStoredCompletionsDataSourceConfig, type EvalCreateResponse as EvalCreateResponse, type EvalRetrieveResponse as EvalRetrieveResponse, type EvalUpdateResponse as EvalUpdateResponse, type EvalListResponse as EvalListResponse, type EvalDeleteResponse as EvalDeleteResponse, EvalListResponsesPage as EvalListResponsesPage, type EvalCreateParams as EvalCreateParams, type EvalUpdateParams as EvalUpdateParams, type EvalListParams as EvalListParams, };
     export { Runs as Runs, type CreateEvalCompletionsRunDataSource as CreateEvalCompletionsRunDataSource, type CreateEvalJSONLRunDataSource as CreateEvalJSONLRunDataSource, type EvalAPIError as EvalAPIError, type RunCreateResponse as RunCreateResponse, type RunRetrieveResponse as RunRetrieveResponse, type RunListResponse as RunListResponse, type RunDeleteResponse as RunDeleteResponse, type RunCancelResponse as RunCancelResponse, RunListResponsesPage as RunListResponsesPage, type RunCreateParams as RunCreateParams, type RunListParams as RunListParams, };
 }
 //# sourceMappingURL=evals.d.ts.map