vellum-ai 0.6.3 → 0.6.6

package/core/streaming-fetcher/Stream.js

@@ -29,37 +29,61 @@ var __asyncGenerator = (this && this.__asyncGenerator) || function (thisArg, _ar
 };
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.readableStreamAsyncIterable = exports.Stream = void 0;
+const runtime_1 = require("../runtime");
+const DATA_PREFIX = "data:";
 class Stream {
-    constructor({ stream, parse, terminator, }) {
+    constructor({ stream, parse, eventShape, signal }) {
+        this.controller = new AbortController();
         this.stream = stream;
         this.parse = parse;
-        this.terminator = terminator;
+        if (eventShape.type === "sse") {
+            this.prefix = DATA_PREFIX;
+            this.messageTerminator = "\n";
+            this.streamTerminator = eventShape.streamTerminator;
+        }
+        else {
+            this.messageTerminator = eventShape.messageTerminator;
+        }
+        signal === null || signal === void 0 ? void 0 : signal.addEventListener("abort", () => this.controller.abort());
     }
     iterMessages() {
         return __asyncGenerator(this, arguments, function* iterMessages_1() {
             var e_1, _a;
-            const decoder = new TextDecoder("utf8");
+            this.controller.signal;
             const stream = readableStreamAsyncIterable(this.stream);
-            let previous = "";
+            let buf = "";
+            let prefixSeen = false;
             try {
                 for (var stream_1 = __asyncValues(stream), stream_1_1; stream_1_1 = yield __await(stream_1.next()), !stream_1_1.done;) {
                     const chunk = stream_1_1.value;
-                    let bufferChunk = "";
-                    // Buffer is present in Node.js environment
-                    if (typeof Buffer !== "undefined") {
-                        bufferChunk += Buffer.isBuffer(chunk) ? chunk : Buffer.from(chunk);
-                    }
-                    // TextDecoder is present in Browser environment
-                    else if (typeof TextDecoder !== "undefined") {
-                        bufferChunk += decoder.decode(chunk);
-                    }
-                    previous += bufferChunk;
+                    buf += this.decodeChunk(chunk);
                     let terminatorIndex;
-                    while ((terminatorIndex = previous.indexOf(this.terminator)) >= 0) {
-                        const line = previous.slice(0, terminatorIndex).trimEnd();
+                    // Parse the chunk into as many messages as possible
+                    while ((terminatorIndex = buf.indexOf(this.messageTerminator)) >= 0) {
+                        // Extract the line from the buffer
+                        let line = buf.slice(0, terminatorIndex + 1);
+                        buf = buf.slice(terminatorIndex + 1);
+                        // Skip empty lines
+                        if (line.length === 0) {
+                            continue;
+                        }
+                        // Skip the chunk until the prefix is found
+                        if (!prefixSeen && this.prefix != null) {
+                            const prefixIndex = line.indexOf(this.prefix);
+                            if (prefixIndex === -1) {
+                                continue;
+                            }
+                            prefixSeen = true;
+                            line = line.slice(prefixIndex + this.prefix.length);
+                        }
+                        // If the stream terminator is present, return
+                        if (this.streamTerminator != null && line.includes(this.streamTerminator)) {
+                            return yield __await(void 0);
+                        }
+                        // Otherwise, yield message from the prefix to the terminator
                         const message = yield __await(this.parse(JSON.parse(line)));
                         yield yield __await(message);
-                        previous = previous.slice(terminatorIndex + 1);
+                        prefixSeen = false;
                     }
                 }
             }
@@ -90,6 +114,19 @@ class Stream {
             }
         });
     }
+    decodeChunk(chunk) {
+        let decoded = "";
+        // If TextDecoder is present, use it
+        if (typeof TextDecoder !== "undefined") {
+            const decoder = new TextDecoder("utf8");
+            decoded += decoder.decode(chunk);
+        }
+        // Buffer is present in Node.js environment
+        else if (runtime_1.RUNTIME.type === "node" && typeof chunk != "undefined") {
+            decoded += Buffer.isBuffer(chunk) ? chunk : Buffer.from(chunk);
+        }
+        return decoded;
+    }
 }
 exports.Stream = Stream;
 /**

package/dist/Client.d.ts

@@ -3,7 +3,7 @@
  */
 import * as environments from "./environments";
 import * as core from "./core";
-import * as Vellum from "./api";
+import * as Vellum from "./api/index";
 import { Deployments } from "./api/resources/deployments/client/Client";
 import { DocumentIndexes } from "./api/resources/documentIndexes/client/Client";
 import { Documents } from "./api/resources/documents/client/Client";
@@ -21,6 +21,7 @@ export declare namespace VellumClient {
     interface RequestOptions {
         timeoutInSeconds?: number;
         maxRetries?: number;
+        abortSignal?: AbortSignal;
     }
 }
 export declare class VellumClient {
@@ -28,10 +29,54 @@ export declare class VellumClient {
     constructor(_options: VellumClient.Options);
     /**
      * Executes a deployed Prompt and returns the result.
+     *
+     * @param {Vellum.ExecutePromptRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.ForbiddenError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
+     *
+     * @example
+     *     await vellum.executePrompt({
+     *         inputs: [{
+     *                 type: "STRING",
+     *                 name: "string",
+     *                 value: "string"
+     *             }],
+     *         promptDeploymentId: "string",
+     *         promptDeploymentName: "string",
+     *         releaseTag: "string",
+     *         externalId: "string",
+     *         expandMeta: {
+     *             modelName: true,
+     *             latency: true,
+     *             deploymentReleaseTag: true,
+     *             promptVersionId: true,
+     *             finishReason: true,
+     *             usage: true
+     *         },
+     *         rawOverrides: {
+     *             body: {
+     *                 "string": {
+     *                     "key": "value"
+     *                 }
+     *             },
+     *             headers: {
+     *                 "string": {
+     *                     "key": "value"
+     *                 }
+     *             },
+     *             url: "string"
+     *         },
+     *         expandRaw: ["string"],
+     *         metadata: {
+     *             "string": {
+     *                 "key": "value"
+     *             }
+     *         }
+     *     })
      */
     executePrompt(request: Vellum.ExecutePromptRequest, requestOptions?: VellumClient.RequestOptions): Promise<Vellum.ExecutePromptResponse>;
     /**
@@ -40,9 +85,26 @@ export declare class VellumClient {
     executePromptStream(request: Vellum.ExecutePromptStreamRequest, requestOptions?: VellumClient.RequestOptions): Promise<core.Stream<Vellum.ExecutePromptEvent>>;
     /**
      * Executes a deployed Workflow and returns its outputs.
+     *
+     * @param {Vellum.ExecuteWorkflowRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
+     *
+     * @example
+     *     await vellum.executeWorkflow({
+     *         inputs: [{
+     *                 type: "STRING",
+     *                 name: "string",
+     *                 value: "string"
+     *             }],
+     *         workflowDeploymentId: "string",
+     *         workflowDeploymentName: "string",
+     *         releaseTag: "string",
+     *         externalId: "string"
+     *     })
      */
     executeWorkflow(request: Vellum.ExecuteWorkflowRequest, requestOptions?: VellumClient.RequestOptions): Promise<Vellum.ExecuteWorkflowResponse>;
     /**
@@ -52,7 +114,12 @@ export declare class VellumClient {
     /**
      * Generate a completion using a previously defined deployment.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * Important: This endpoint is DEPRECATED and has been superseded by
+     * [execute-prompt](/api-reference/api-reference/execute-prompt).
+     *
+     * @param {Vellum.GenerateBodyRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.ForbiddenError}
      * @throws {@link Vellum.NotFoundError}
@@ -69,13 +136,16 @@ export declare class VellumClient {
     /**
      * Generate a stream of completions using a previously defined deployment.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * Important: This endpoint is DEPRECATED and has been superseded by
+     * [execute-prompt-stream](/api-reference/api-reference/execute-prompt-stream).
      */
     generateStream(request: Vellum.GenerateStreamBodyRequest, requestOptions?: VellumClient.RequestOptions): Promise<core.Stream<Vellum.GenerateStreamResponse>>;
     /**
      * Perform a search against a document index.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * @param {Vellum.SearchRequestBodyRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
@@ -89,7 +159,9 @@ export declare class VellumClient {
     /**
      * Used to submit feedback regarding the quality of previously generated completions.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * @param {Vellum.SubmitCompletionActualsRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
@@ -105,6 +177,9 @@ export declare class VellumClient {
      *
      *     **Note:** Uses a base url of `https://predict.vellum.ai`.
      *
+     * @param {Vellum.SubmitWorkflowExecutionActualsRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @example
      *     await vellum.submitWorkflowExecutionActuals({
      *         actuals: []

package/dist/Client.js

@@ -41,10 +41,10 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.VellumClient = void 0;
 const environments = __importStar(require("./environments"));
 const core = __importStar(require("./core"));
-const Vellum = __importStar(require("./api"));
-const serializers = __importStar(require("./serialization"));
+const Vellum = __importStar(require("./api/index"));
+const serializers = __importStar(require("./serialization/index"));
 const url_join_1 = __importDefault(require("url-join"));
-const errors = __importStar(require("./errors"));
+const errors = __importStar(require("./errors/index"));
 const Client_1 = require("./api/resources/deployments/client/Client");
 const Client_2 = require("./api/resources/documentIndexes/client/Client");
 const Client_3 = require("./api/resources/documents/client/Client");
@@ -60,10 +60,54 @@ class VellumClient {
     }
     /**
      * Executes a deployed Prompt and returns the result.
+     *
+     * @param {Vellum.ExecutePromptRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.ForbiddenError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
+     *
+     * @example
+     *     await vellum.executePrompt({
+     *         inputs: [{
+     *                 type: "STRING",
+     *                 name: "string",
+     *                 value: "string"
+     *             }],
+     *         promptDeploymentId: "string",
+     *         promptDeploymentName: "string",
+     *         releaseTag: "string",
+     *         externalId: "string",
+     *         expandMeta: {
+     *             modelName: true,
+     *             latency: true,
+     *             deploymentReleaseTag: true,
+     *             promptVersionId: true,
+     *             finishReason: true,
+     *             usage: true
+     *         },
+     *         rawOverrides: {
+     *             body: {
+     *                 "string": {
+     *                     "key": "value"
+     *                 }
+     *             },
+     *             headers: {
+     *                 "string": {
+     *                     "key": "value"
+     *                 }
+     *             },
+     *             url: "string"
+     *         },
+     *         expandRaw: ["string"],
+     *         metadata: {
+     *             "string": {
+     *                 "key": "value"
+     *             }
+     *         }
+     *     })
      */
     executePrompt(request, requestOptions) {
         var _a;
@@ -72,11 +116,12 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/execute-prompt"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.ExecutePromptRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return yield serializers.ExecutePromptResponse.parseOrThrow(_response.body, {
@@ -128,7 +173,7 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/execute-prompt-stream"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.ExecutePromptStreamRequest.jsonOrThrow(request, {
                     unrecognizedObjectKeys: "strip",
@@ -136,11 +181,11 @@ class VellumClient {
                 responseType: "streaming",
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return new core.Stream({
                     stream: _response.body,
-                    terminator: "\n",
                     parse: (data) => __awaiter(this, void 0, void 0, function* () {
                         return yield serializers.ExecutePromptEvent.parseOrThrow(data, {
                             unrecognizedObjectKeys: "passthrough",
@@ -149,6 +194,11 @@ class VellumClient {
                             breadcrumbsPrefix: ["response"],
                         });
                     }),
+                    signal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
+                    eventShape: {
+                        type: "json",
+                        messageTerminator: "\n",
+                    },
                 });
             }
             if (_response.error.reason === "status-code") {
@@ -185,9 +235,26 @@ class VellumClient {
     }
     /**
      * Executes a deployed Workflow and returns its outputs.
+     *
+     * @param {Vellum.ExecuteWorkflowRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
+     *
+     * @example
+     *     await vellum.executeWorkflow({
+     *         inputs: [{
+     *                 type: "STRING",
+     *                 name: "string",
+     *                 value: "string"
+     *             }],
+     *         workflowDeploymentId: "string",
+     *         workflowDeploymentName: "string",
+     *         releaseTag: "string",
+     *         externalId: "string"
+     *     })
      */
     executeWorkflow(request, requestOptions) {
         var _a;
@@ -196,11 +263,12 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/execute-workflow"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.ExecuteWorkflowRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return yield serializers.ExecuteWorkflowResponse.parseOrThrow(_response.body, {
@@ -250,7 +318,7 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/execute-workflow-stream"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.ExecuteWorkflowStreamRequest.jsonOrThrow(request, {
                     unrecognizedObjectKeys: "strip",
@@ -258,11 +326,11 @@ class VellumClient {
                 responseType: "streaming",
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return new core.Stream({
                     stream: _response.body,
-                    terminator: "\n",
                     parse: (data) => __awaiter(this, void 0, void 0, function* () {
                         return yield serializers.WorkflowStreamEvent.parseOrThrow(data, {
                             unrecognizedObjectKeys: "passthrough",
@@ -271,6 +339,11 @@ class VellumClient {
                             breadcrumbsPrefix: ["response"],
                         });
                     }),
+                    signal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
+                    eventShape: {
+                        type: "json",
+                        messageTerminator: "\n",
+                    },
                 });
             }
             if (_response.error.reason === "status-code") {
@@ -306,7 +379,12 @@ class VellumClient {
     /**
      * Generate a completion using a previously defined deployment.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * Important: This endpoint is DEPRECATED and has been superseded by
+     * [execute-prompt](/api-reference/api-reference/execute-prompt).
+     *
+     * @param {Vellum.GenerateBodyRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.ForbiddenError}
      * @throws {@link Vellum.NotFoundError}
@@ -326,11 +404,12 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/generate"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.GenerateBodyRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return yield serializers.GenerateResponse.parseOrThrow(_response.body, {
@@ -375,7 +454,8 @@ class VellumClient {
     /**
      * Generate a stream of completions using a previously defined deployment.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * Important: This endpoint is DEPRECATED and has been superseded by
+     * [execute-prompt-stream](/api-reference/api-reference/execute-prompt-stream).
      */
     generateStream(request, requestOptions) {
         var _a;
@@ -384,17 +464,17 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/generate-stream"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.GenerateStreamBodyRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
                 responseType: "streaming",
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return new core.Stream({
                     stream: _response.body,
-                    terminator: "\n",
                     parse: (data) => __awaiter(this, void 0, void 0, function* () {
                         return yield serializers.GenerateStreamResponse.parseOrThrow(data, {
                             unrecognizedObjectKeys: "passthrough",
@@ -403,6 +483,11 @@ class VellumClient {
                             breadcrumbsPrefix: ["response"],
                         });
                     }),
+                    signal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
+                    eventShape: {
+                        type: "json",
+                        messageTerminator: "\n",
+                    },
                 });
             }
             if (_response.error.reason === "status-code") {
@@ -440,7 +525,9 @@ class VellumClient {
     /**
      * Perform a search against a document index.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * @param {Vellum.SearchRequestBodyRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
@@ -457,11 +544,12 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/search"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.SearchRequestBodyRequest.jsonOrThrow(request, { unrecognizedObjectKeys: "strip" }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return yield serializers.SearchResponse.parseOrThrow(_response.body, {
@@ -504,7 +592,9 @@ class VellumClient {
     /**
      * Used to submit feedback regarding the quality of previously generated completions.
      *
-     * **Note:** Uses a base url of `https://predict.vellum.ai`.
+     * @param {Vellum.SubmitCompletionActualsRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @throws {@link Vellum.BadRequestError}
      * @throws {@link Vellum.NotFoundError}
      * @throws {@link Vellum.InternalServerError}
@@ -521,13 +611,14 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/submit-completion-actuals"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.SubmitCompletionActualsRequest.jsonOrThrow(request, {
                     unrecognizedObjectKeys: "strip",
                 }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return;
@@ -567,6 +658,9 @@ class VellumClient {
      *
      *     **Note:** Uses a base url of `https://predict.vellum.ai`.
      *
+     * @param {Vellum.SubmitWorkflowExecutionActualsRequest} request
+     * @param {VellumClient.RequestOptions} requestOptions - Request-specific configuration.
+     *
      * @example
      *     await vellum.submitWorkflowExecutionActuals({
      *         actuals: []
@@ -579,13 +673,14 @@ class VellumClient {
                 url: (0, url_join_1.default)(((_a = (yield core.Supplier.get(this._options.environment))) !== null && _a !== void 0 ? _a : environments.VellumEnvironment.Production)
                     .predict, "v1/submit-workflow-execution-actuals"),
                 method: "POST",
-                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.3", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
+                headers: Object.assign({ "X-Fern-Language": "JavaScript", "X-Fern-SDK-Name": "vellum-ai", "X-Fern-SDK-Version": "0.6.6", "X-Fern-Runtime": core.RUNTIME.type, "X-Fern-Runtime-Version": core.RUNTIME.version }, (yield this._getCustomAuthorizationHeaders())),
                 contentType: "application/json",
                 body: yield serializers.SubmitWorkflowExecutionActualsRequest.jsonOrThrow(request, {
                     unrecognizedObjectKeys: "strip",
                 }),
                 timeoutMs: (requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.timeoutInSeconds) != null ? requestOptions.timeoutInSeconds * 1000 : undefined,
                 maxRetries: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.maxRetries,
+                abortSignal: requestOptions === null || requestOptions === void 0 ? void 0 : requestOptions.abortSignal,
             });
             if (_response.ok) {
                 return;

package/dist/api/client/requests/ExecutePromptRequest.d.ts

@@ -1,7 +1,48 @@
 /**
  * This file was auto-generated by Fern from our API Definition.
  */
-import * as Vellum from "../..";
+import * as Vellum from "../../index";
+/**
+ * @example
+ *     {
+ *         inputs: [{
+ *                 type: "STRING",
+ *                 name: "string",
+ *                 value: "string"
+ *             }],
+ *         promptDeploymentId: "string",
+ *         promptDeploymentName: "string",
+ *         releaseTag: "string",
+ *         externalId: "string",
+ *         expandMeta: {
+ *             modelName: true,
+ *             latency: true,
+ *             deploymentReleaseTag: true,
+ *             promptVersionId: true,
+ *             finishReason: true,
+ *             usage: true
+ *         },
+ *         rawOverrides: {
+ *             body: {
+ *                 "string": {
+ *                     "key": "value"
+ *                 }
+ *             },
+ *             headers: {
+ *                 "string": {
+ *                     "key": "value"
+ *                 }
+ *             },
+ *             url: "string"
+ *         },
+ *         expandRaw: ["string"],
+ *         metadata: {
+ *             "string": {
+ *                 "key": "value"
+ *             }
+ *         }
+ *     }
+ */
 export interface ExecutePromptRequest {
     /** The list of inputs defined in the Prompt's deployment with their corresponding values. */
     inputs: Vellum.PromptDeploymentInputRequest[];