langchain 0.0.175 → 0.0.177

package/dist/chat_models/bedrock.cjs

@@ -133,6 +133,12 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
             writable: true,
             value: new eventstream_codec_1.EventStreamCodec(util_utf8_1.toUtf8, util_utf8_1.fromUtf8)
         });
+        Object.defineProperty(this, "streaming", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: false
+        });
         this.model = fields?.model ?? this.model;
         const allowedModels = ["ai21", "anthropic", "amazon"];
         if (!allowedModels.includes(this.model.split(".")[0])) {
@@ -150,6 +156,7 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
         this.endpointHost = fields?.endpointHost ?? fields?.endpointUrl;
         this.stopSequences = fields?.stopSequences;
         this.modelKwargs = fields?.modelKwargs;
+        this.streaming = fields?.streaming ?? this.streaming;
     }
     /** Call out to Bedrock service model.
       Arguments:
@@ -161,10 +168,23 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
       Example:
         response = model.call("Tell me a joke.")
     */
-    async _call(messages, options) {
+    async _call(messages, options, runManager) {
         const service = "bedrock-runtime";
         const endpointHost = this.endpointHost ?? `${service}.${this.region}.amazonaws.com`;
         const provider = this.model.split(".")[0];
+        if (this.streaming) {
+            const stream = this._streamResponseChunks(messages, options, runManager);
+            let finalResult;
+            for await (const chunk of stream) {
+                if (finalResult === undefined) {
+                    finalResult = chunk;
+                }
+                else {
+                    finalResult = finalResult.concat(chunk);
+                }
+            }
+            return finalResult?.message.content ?? "";
+        }
         const response = await this._signedFetch(messages, options, {
             bedrockMethod: "invoke",
             endpointHost,
@@ -233,7 +253,6 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
                     event.headers[":content-type"].value !== "application/json") {
                     throw Error(`Failed to get event chunk: got ${chunk}`);
                 }
-                // console.log(decoder.decode(event.body));
                 const body = JSON.parse(decoder.decode(event.body));
                 if (body.message) {
                     throw new Error(body.message);
@@ -245,7 +264,8 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
                         text,
                         message: new index_js_1.AIMessageChunk({ content: text }),
                     });
-                    await runManager?.handleLLMNewToken(text);
+                    // eslint-disable-next-line no-void
+                    void runManager?.handleLLMNewToken(text);
                 }
             }
         }
@@ -256,7 +276,8 @@ class ChatBedrock extends base_js_1.SimpleChatModel {
                 text,
                 message: new index_js_1.AIMessageChunk({ content: text }),
             });
-            await runManager?.handleLLMNewToken(text);
+            // eslint-disable-next-line no-void
+            void runManager?.handleLLMNewToken(text);
         }
     }
     // eslint-disable-next-line @typescript-eslint/no-explicit-any

package/dist/chat_models/bedrock.d.ts

@@ -33,6 +33,7 @@ export declare class ChatBedrock extends SimpleChatModel implements BaseBedrockI
     stopSequences?: string[];
     modelKwargs?: Record<string, unknown>;
     codec: EventStreamCodec;
+    streaming: boolean;
     get lc_secrets(): {
         [key: string]: string;
     } | undefined;
@@ -49,7 +50,7 @@ export declare class ChatBedrock extends SimpleChatModel implements BaseBedrockI
       Example:
         response = model.call("Tell me a joke.")
     */
-    _call(messages: BaseMessage[], options: this["ParsedCallOptions"]): Promise<string>;
+    _call(messages: BaseMessage[], options: this["ParsedCallOptions"], runManager?: CallbackManagerForLLMRun): Promise<string>;
     _signedFetch(messages: BaseMessage[], options: this["ParsedCallOptions"], fields: {
         bedrockMethod: "invoke" | "invoke-with-response-stream";
         endpointHost: string;

package/dist/chat_models/bedrock.js

@@ -128,6 +128,12 @@ export class ChatBedrock extends SimpleChatModel {
             writable: true,
             value: new EventStreamCodec(toUtf8, fromUtf8)
         });
+        Object.defineProperty(this, "streaming", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: false
+        });
         this.model = fields?.model ?? this.model;
         const allowedModels = ["ai21", "anthropic", "amazon"];
         if (!allowedModels.includes(this.model.split(".")[0])) {
@@ -145,6 +151,7 @@ export class ChatBedrock extends SimpleChatModel {
         this.endpointHost = fields?.endpointHost ?? fields?.endpointUrl;
         this.stopSequences = fields?.stopSequences;
         this.modelKwargs = fields?.modelKwargs;
+        this.streaming = fields?.streaming ?? this.streaming;
     }
     /** Call out to Bedrock service model.
       Arguments:
@@ -156,10 +163,23 @@ export class ChatBedrock extends SimpleChatModel {
       Example:
         response = model.call("Tell me a joke.")
     */
-    async _call(messages, options) {
+    async _call(messages, options, runManager) {
         const service = "bedrock-runtime";
         const endpointHost = this.endpointHost ?? `${service}.${this.region}.amazonaws.com`;
         const provider = this.model.split(".")[0];
+        if (this.streaming) {
+            const stream = this._streamResponseChunks(messages, options, runManager);
+            let finalResult;
+            for await (const chunk of stream) {
+                if (finalResult === undefined) {
+                    finalResult = chunk;
+                }
+                else {
+                    finalResult = finalResult.concat(chunk);
+                }
+            }
+            return finalResult?.message.content ?? "";
+        }
         const response = await this._signedFetch(messages, options, {
             bedrockMethod: "invoke",
             endpointHost,
@@ -228,7 +248,6 @@ export class ChatBedrock extends SimpleChatModel {
                     event.headers[":content-type"].value !== "application/json") {
                     throw Error(`Failed to get event chunk: got ${chunk}`);
                 }
-                // console.log(decoder.decode(event.body));
                 const body = JSON.parse(decoder.decode(event.body));
                 if (body.message) {
                     throw new Error(body.message);
@@ -240,7 +259,8 @@ export class ChatBedrock extends SimpleChatModel {
                         text,
                         message: new AIMessageChunk({ content: text }),
                     });
-                    await runManager?.handleLLMNewToken(text);
+                    // eslint-disable-next-line no-void
+                    void runManager?.handleLLMNewToken(text);
                 }
             }
         }
@@ -251,7 +271,8 @@ export class ChatBedrock extends SimpleChatModel {
                 text,
                 message: new AIMessageChunk({ content: text }),
             });
-            await runManager?.handleLLMNewToken(text);
+            // eslint-disable-next-line no-void
+            void runManager?.handleLLMNewToken(text);
         }
     }
     // eslint-disable-next-line @typescript-eslint/no-explicit-any

package/dist/chat_models/googlevertexai/common.cjs

@@ -144,6 +144,12 @@ class BaseChatGoogleVertexAI extends base_js_1.BaseChatModel {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "streamedConnection", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
         this.model = fields?.model ?? this.model;
         this.temperature = fields?.temperature ?? this.temperature;
         this.maxOutputTokens = fields?.maxOutputTokens ?? this.maxOutputTokens;
@@ -155,15 +161,31 @@ class BaseChatGoogleVertexAI extends base_js_1.BaseChatModel {
         // TODO: Combine the safetyAttributes
         return [];
     }
-    // TODO: Add streaming support
+    async *_streamResponseChunks(_messages, _options, _runManager) {
+        // Make the call as a streaming request
+        const instance = this.createInstance(_messages);
+        const parameters = this.formatParameters();
+        const result = await this.streamedConnection.request([instance], parameters, _options);
+        // Get the streaming parser of the response
+        const stream = result.data;
+        // Loop until the end of the stream
+        // During the loop, yield each time we get a chunk from the streaming parser
+        // that is either available or added to the queue
+        while (!stream.streamDone) {
+            const output = await stream.nextChunk();
+            const chunk = output !== null
+                ? BaseChatGoogleVertexAI.convertPredictionChunk(output)
+                : new index_js_1.ChatGenerationChunk({
+                    text: "",
+                    message: new index_js_1.AIMessageChunk(""),
+                    generationInfo: { finishReason: "stop" },
+                });
+            yield chunk;
+        }
+    }
     async _generate(messages, options) {
         const instance = this.createInstance(messages);
-        const parameters = {
-            temperature: this.temperature,
-            topK: this.topK,
-            topP: this.topP,
-            maxOutputTokens: this.maxOutputTokens,
-        };
+        const parameters = this.formatParameters();
         const result = await this.connection.request([instance], parameters, options);
         const generations = result?.data?.predictions?.map((prediction) => BaseChatGoogleVertexAI.convertPrediction(prediction)) ?? [];
         return {
@@ -211,6 +233,14 @@ class BaseChatGoogleVertexAI extends base_js_1.BaseChatModel {
         };
         return instance;
     }
+    formatParameters() {
+        return {
+            temperature: this.temperature,
+            topK: this.topK,
+            topP: this.topP,
+            maxOutputTokens: this.maxOutputTokens,
+        };
+    }
     /**
      * Converts a prediction from the Google Vertex AI chat model to a chat
      * generation.
@@ -225,5 +255,14 @@ class BaseChatGoogleVertexAI extends base_js_1.BaseChatModel {
             generationInfo: prediction,
         };
     }
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    static convertPredictionChunk(output) {
+        const generation = BaseChatGoogleVertexAI.convertPrediction(output.outputs[0]);
+        return new index_js_1.ChatGenerationChunk({
+            text: generation.text,
+            message: new index_js_1.AIMessageChunk(generation.message),
+            generationInfo: generation.generationInfo,
+        });
+    }
 }
 exports.BaseChatGoogleVertexAI = BaseChatGoogleVertexAI;

package/dist/chat_models/googlevertexai/common.d.ts

@@ -1,8 +1,9 @@
 import { BaseChatModel } from "../base.js";
-import { BaseMessage, ChatGeneration, ChatMessage, ChatResult, LLMResult } from "../../schema/index.js";
+import { BaseMessage, ChatGeneration, ChatGenerationChunk, ChatMessage, ChatResult, LLMResult } from "../../schema/index.js";
 import { GoogleVertexAILLMConnection } from "../../util/googlevertexai-connection.js";
-import { GoogleVertexAIBaseLLMInput, GoogleVertexAIBasePrediction } from "../../types/googlevertexai-types.js";
+import { GoogleVertexAIBaseLLMInput, GoogleVertexAIBasePrediction, GoogleVertexAIModelParams } from "../../types/googlevertexai-types.js";
 import { BaseLanguageModelCallOptions } from "../../base_language/index.js";
+import { CallbackManagerForLLMRun } from "../../callbacks/index.js";
 /**
  * Represents a single "example" exchange that can be provided to
  * help illustrate what a model response should look like.
@@ -96,9 +97,11 @@ export declare class BaseChatGoogleVertexAI<AuthOptions> extends BaseChatModel i
     topK: number;
     examples: ChatExample[];
     connection: GoogleVertexAILLMConnection<BaseLanguageModelCallOptions, GoogleVertexAIChatInstance, GoogleVertexAIChatPrediction, AuthOptions>;
+    streamedConnection: GoogleVertexAILLMConnection<BaseLanguageModelCallOptions, GoogleVertexAIChatInstance, GoogleVertexAIChatPrediction, AuthOptions>;
     get lc_aliases(): Record<string, string>;
     constructor(fields?: GoogleVertexAIChatInput<AuthOptions>);
     _combineLLMOutput(): LLMResult["llmOutput"];
+    _streamResponseChunks(_messages: BaseMessage[], _options: this["ParsedCallOptions"], _runManager?: CallbackManagerForLLMRun): AsyncGenerator<ChatGenerationChunk>;
     _generate(messages: BaseMessage[], options: this["ParsedCallOptions"]): Promise<ChatResult>;
     _llmType(): string;
     /**
@@ -107,6 +110,7 @@ export declare class BaseChatGoogleVertexAI<AuthOptions> extends BaseChatModel i
      * @returns A new instance of the Google Vertex AI chat model.
      */
     createInstance(messages: BaseMessage[]): GoogleVertexAIChatInstance;
+    formatParameters(): GoogleVertexAIModelParams;
     /**
      * Converts a prediction from the Google Vertex AI chat model to a chat
      * generation.
@@ -114,5 +118,6 @@ export declare class BaseChatGoogleVertexAI<AuthOptions> extends BaseChatModel i
      * @returns The converted chat generation.
      */
     static convertPrediction(prediction: GoogleVertexAIChatPrediction): ChatGeneration;
+    static convertPredictionChunk(output: any): ChatGenerationChunk;
 }
 export {};

package/dist/chat_models/googlevertexai/common.js

@@ -1,5 +1,5 @@
 import { BaseChatModel } from "../base.js";
-import { AIMessage, ChatMessage, } from "../../schema/index.js";
+import { AIMessage, AIMessageChunk, ChatGenerationChunk, ChatMessage, } from "../../schema/index.js";
 /**
  * Represents a chat message in the Google Vertex AI chat model.
  */
@@ -140,6 +140,12 @@ export class BaseChatGoogleVertexAI extends BaseChatModel {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "streamedConnection", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: void 0
+        });
         this.model = fields?.model ?? this.model;
         this.temperature = fields?.temperature ?? this.temperature;
         this.maxOutputTokens = fields?.maxOutputTokens ?? this.maxOutputTokens;
@@ -151,15 +157,31 @@ export class BaseChatGoogleVertexAI extends BaseChatModel {
         // TODO: Combine the safetyAttributes
         return [];
     }
-    // TODO: Add streaming support
+    async *_streamResponseChunks(_messages, _options, _runManager) {
+        // Make the call as a streaming request
+        const instance = this.createInstance(_messages);
+        const parameters = this.formatParameters();
+        const result = await this.streamedConnection.request([instance], parameters, _options);
+        // Get the streaming parser of the response
+        const stream = result.data;
+        // Loop until the end of the stream
+        // During the loop, yield each time we get a chunk from the streaming parser
+        // that is either available or added to the queue
+        while (!stream.streamDone) {
+            const output = await stream.nextChunk();
+            const chunk = output !== null
+                ? BaseChatGoogleVertexAI.convertPredictionChunk(output)
+                : new ChatGenerationChunk({
+                    text: "",
+                    message: new AIMessageChunk(""),
+                    generationInfo: { finishReason: "stop" },
+                });
+            yield chunk;
+        }
+    }
     async _generate(messages, options) {
         const instance = this.createInstance(messages);
-        const parameters = {
-            temperature: this.temperature,
-            topK: this.topK,
-            topP: this.topP,
-            maxOutputTokens: this.maxOutputTokens,
-        };
+        const parameters = this.formatParameters();
         const result = await this.connection.request([instance], parameters, options);
         const generations = result?.data?.predictions?.map((prediction) => BaseChatGoogleVertexAI.convertPrediction(prediction)) ?? [];
         return {
@@ -207,6 +229,14 @@ export class BaseChatGoogleVertexAI extends BaseChatModel {
         };
         return instance;
     }
+    formatParameters() {
+        return {
+            temperature: this.temperature,
+            topK: this.topK,
+            topP: this.topP,
+            maxOutputTokens: this.maxOutputTokens,
+        };
+    }
     /**
      * Converts a prediction from the Google Vertex AI chat model to a chat
      * generation.
@@ -221,4 +251,13 @@ export class BaseChatGoogleVertexAI extends BaseChatModel {
             generationInfo: prediction,
         };
     }
+    // eslint-disable-next-line @typescript-eslint/no-explicit-any
+    static convertPredictionChunk(output) {
+        const generation = BaseChatGoogleVertexAI.convertPrediction(output.outputs[0]);
+        return new ChatGenerationChunk({
+            text: generation.text,
+            message: new AIMessageChunk(generation.message),
+            generationInfo: generation.generationInfo,
+        });
+    }
 }

package/dist/chat_models/googlevertexai/index.cjs

@@ -1,9 +1,9 @@
 "use strict";
 Object.defineProperty(exports, "__esModule", { value: true });
 exports.ChatGoogleVertexAI = void 0;
-const google_auth_library_1 = require("google-auth-library");
 const common_js_1 = require("./common.cjs");
 const googlevertexai_connection_js_1 = require("../../util/googlevertexai-connection.cjs");
+const googlevertexai_gauth_js_1 = require("../../util/googlevertexai-gauth.cjs");
 /**
  * Enables calls to the Google Cloud's Vertex AI API to access
  * Large Language Models in a chat-like fashion.
@@ -24,11 +24,12 @@ class ChatGoogleVertexAI extends common_js_1.BaseChatGoogleVertexAI {
     }
     constructor(fields) {
         super(fields);
-        const client = new google_auth_library_1.GoogleAuth({
+        const client = new googlevertexai_gauth_js_1.GAuthClient({
             scopes: "https://www.googleapis.com/auth/cloud-platform",
             ...fields?.authOptions,
         });
-        this.connection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client);
+        this.connection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, false);
+        this.streamedConnection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, true);
     }
 }
 exports.ChatGoogleVertexAI = ChatGoogleVertexAI;

package/dist/chat_models/googlevertexai/index.js

@@ -1,6 +1,6 @@
-import { GoogleAuth } from "google-auth-library";
 import { BaseChatGoogleVertexAI } from "./common.js";
 import { GoogleVertexAILLMConnection } from "../../util/googlevertexai-connection.js";
+import { GAuthClient } from "../../util/googlevertexai-gauth.js";
 /**
  * Enables calls to the Google Cloud's Vertex AI API to access
  * Large Language Models in a chat-like fashion.
@@ -21,10 +21,11 @@ export class ChatGoogleVertexAI extends BaseChatGoogleVertexAI {
     }
     constructor(fields) {
         super(fields);
-        const client = new GoogleAuth({
+        const client = new GAuthClient({
             scopes: "https://www.googleapis.com/auth/cloud-platform",
             ...fields?.authOptions,
         });
-        this.connection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client);
+        this.connection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, false);
+        this.streamedConnection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, true);
     }
 }

package/dist/chat_models/googlevertexai/web.cjs

@@ -25,7 +25,8 @@ class ChatGoogleVertexAI extends common_js_1.BaseChatGoogleVertexAI {
     constructor(fields) {
         super(fields);
         const client = new googlevertexai_webauth_js_1.WebGoogleAuth(fields?.authOptions);
-        this.connection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client);
+        this.connection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, false);
+        this.streamedConnection = new googlevertexai_connection_js_1.GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, true);
     }
 }
 exports.ChatGoogleVertexAI = ChatGoogleVertexAI;

package/dist/chat_models/googlevertexai/web.js

@@ -22,6 +22,7 @@ export class ChatGoogleVertexAI extends BaseChatGoogleVertexAI {
     constructor(fields) {
         super(fields);
         const client = new WebGoogleAuth(fields?.authOptions);
-        this.connection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client);
+        this.connection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, false);
+        this.streamedConnection = new GoogleVertexAILLMConnection({ ...fields, ...this }, this.caller, client, true);
     }
 }

package/dist/chat_models/llama_cpp.cjs

@@ -3,6 +3,7 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.ChatLlamaCpp = void 0;
 const node_llama_cpp_1 = require("node-llama-cpp");
 const base_js_1 = require("./base.cjs");
+const llama_cpp_js_1 = require("../util/llama_cpp.cjs");
 /**
  *  To use this model you need to have the `node-llama-cpp` module installed.
  *  This can be installed using `npm install -S node-llama-cpp` and the minimum
@@ -15,73 +16,31 @@ class ChatLlamaCpp extends base_js_1.SimpleChatModel {
     }
     constructor(inputs) {
         super(inputs);
-        Object.defineProperty(this, "batchSize", {
+        Object.defineProperty(this, "maxTokens", {
             enumerable: true,
             configurable: true,
             writable: true,
             value: void 0
         });
-        Object.defineProperty(this, "contextSize", {
+        Object.defineProperty(this, "temperature", {
             enumerable: true,
             configurable: true,
             writable: true,
             value: void 0
         });
-        Object.defineProperty(this, "embedding", {
+        Object.defineProperty(this, "topK", {
             enumerable: true,
             configurable: true,
             writable: true,
             value: void 0
         });
-        Object.defineProperty(this, "f16Kv", {
+        Object.defineProperty(this, "topP", {
             enumerable: true,
             configurable: true,
             writable: true,
             value: void 0
         });
-        Object.defineProperty(this, "gpuLayers", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "logitsAll", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "lowVram", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "seed", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "useMlock", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "useMmap", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "vocabOnly", {
-            enumerable: true,
-            configurable: true,
-            writable: true,
-            value: void 0
-        });
-        Object.defineProperty(this, "modelPath", {
+        Object.defineProperty(this, "trimWhitespaceSuffix", {
             enumerable: true,
             configurable: true,
             writable: true,
@@ -105,47 +64,33 @@ class ChatLlamaCpp extends base_js_1.SimpleChatModel {
             writable: true,
             value: void 0
         });
-        this.batchSize = inputs?.batchSize;
-        this.contextSize = inputs?.contextSize;
-        this.embedding = inputs?.embedding;
-        this.f16Kv = inputs?.f16Kv;
-        this.gpuLayers = inputs?.gpuLayers;
-        this.logitsAll = inputs?.logitsAll;
-        this.lowVram = inputs?.lowVram;
-        this.modelPath = inputs.modelPath;
-        this.seed = inputs?.seed;
-        this.useMlock = inputs?.useMlock;
-        this.useMmap = inputs?.useMmap;
-        this.vocabOnly = inputs?.vocabOnly;
-        this._model = new node_llama_cpp_1.LlamaModel(inputs);
-        this._context = new node_llama_cpp_1.LlamaContext({ model: this._model });
+        this.maxTokens = inputs?.maxTokens;
+        this.temperature = inputs?.temperature;
+        this.topK = inputs?.topK;
+        this.topP = inputs?.topP;
+        this.trimWhitespaceSuffix = inputs?.trimWhitespaceSuffix;
+        this._model = (0, llama_cpp_js_1.createLlamaModel)(inputs);
+        this._context = (0, llama_cpp_js_1.createLlamaContext)(this._model, inputs);
         this._session = null;
     }
     _llmType() {
         return "llama2_cpp";
     }
-    invocationParams() {
-        return {
-            batchSize: this.batchSize,
-            contextSize: this.contextSize,
-            embedding: this.embedding,
-            f16Kv: this.f16Kv,
-            gpuLayers: this.gpuLayers,
-            logitsAll: this.logitsAll,
-            lowVram: this.lowVram,
-            modelPath: this.modelPath,
-            seed: this.seed,
-            useMlock: this.useMlock,
-            useMmap: this.useMmap,
-            vocabOnly: this.vocabOnly,
-        };
-    }
     /** @ignore */
     _combineLLMOutput() {
         return {};
     }
+    invocationParams() {
+        return {
+            maxTokens: this.maxTokens,
+            temperature: this.temperature,
+            topK: this.topK,
+            topP: this.topP,
+            trimWhitespaceSuffix: this.trimWhitespaceSuffix,
+        };
+    }
     /** @ignore */
-    async _call(messages, options) {
+    async _call(messages, _options) {
         let prompt = "";
         if (messages.length > 1) {
             // We need to build a new _session
@@ -159,8 +104,15 @@ class ChatLlamaCpp extends base_js_1.SimpleChatModel {
             prompt = messages[0].content;
         }
         try {
+            const promptOptions = {
+                maxTokens: this?.maxTokens,
+                temperature: this?.temperature,
+                topK: this?.topK,
+                topP: this?.topP,
+                trimWhitespaceSuffix: this?.trimWhitespaceSuffix,
+            };
             // @ts-expect-error - TS2531: Object is possibly 'null'.
-            const completion = await this._session.prompt(prompt, options);
+            const completion = await this._session.prompt(prompt, promptOptions);
             return completion;
         }
         catch (e) {