modelfusion 0.70.0 → 0.72.0

package/README.md

@@ -15,7 +15,7 @@
 **ModelFusion** is a TypeScript library for building AI applications, chatbots, and agents.
 
 - **Vendor-neutral**: ModelFusion is a non-commercial open source project that is community-driven. You can use it with any supported provider.
-- **Multimodal**: ModelFusion supports a wide range of models including text generation, image generation, vision, text-to-speech, speech-to-text, and embedding models.
+- **Multi-modal**: ModelFusion supports a wide range of models including text generation, image generation, vision, text-to-speech, speech-to-text, and embedding models.
 - **Streaming**: ModelFusion supports streaming for many generation models, e.g. text streaming, structure streaming, and full duplex speech streaming.
 - **Utility functions**: ModelFusion provides functionality for tools and tool usage, vector indices, and guards functions.
 - **Type inference and validation**: ModelFusion infers TypeScript types wherever possible and to validates model responses.
@@ -52,7 +52,7 @@ const text = await generateText(
 );
 ```
 
-Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Anthropic](https://modelfusion.dev/integration/model-provider/anthropic), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama), [Hugging Face](https://modelfusion.dev/integration/model-provider/huggingface)
+Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama), [Hugging Face](https://modelfusion.dev/integration/model-provider/huggingface), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Anthropic](https://modelfusion.dev/integration/model-provider/anthropic)
 
 #### streamText
 
@@ -67,7 +67,7 @@ for await (const textPart of textStream) {
 }
 ```
 
-Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Anthropic](https://modelfusion.dev/integration/model-provider/anthropic), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama)
+Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Anthropic](https://modelfusion.dev/integration/model-provider/anthropic)
 
 #### streamText with multi-modal prompt
 
@@ -250,39 +250,6 @@ for await (const part of structureStream) {
 
 Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai)
 
-### [Generate Structure or Text](https://modelfusion.dev/guide/function/generate-structure-or-text)
-
-Generate a structure (or text as a fallback) using a prompt and multiple schemas.
-It either matches one of the schemas or is text reponse.
-
-```ts
-const { structure, value, text } = await generateStructureOrText(
-  new OpenAIChatModel({ model: "gpt-3.5-turbo", maxCompletionTokens: 1000 }),
-  [
-    new ZodStructureDefinition({
-      name: "getCurrentWeather" as const, // mark 'as const' for type inference
-      description: "Get the current weather in a given location",
-      schema: z.object({
-        location: z
-          .string()
-          .describe("The city and state, e.g. San Francisco, CA"),
-        unit: z.enum(["celsius", "fahrenheit"]).optional(),
-      }),
-    }),
-    new ZodStructureDefinition({
-      name: "getContactInformation" as const,
-      description: "Get the contact information for a given person",
-      schema: z.object({
-        name: z.string().describe("The name of the person"),
-      }),
-    }),
-  ],
-  [OpenAIChatMessage.user(query)]
-);
-```
-
-Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai)
-
 ### [Embed Value](https://modelfusion.dev/guide/function/embed)
 
 Create embeddings for text and other values. Embeddings are vectors that represent the essence of the values in the context of the model.
@@ -304,7 +271,7 @@ const embeddings = await embedMany(
 );
 ```
 
-Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama), [Hugging Face](https://modelfusion.dev/integration/model-provider/huggingface)
+Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Ollama](https://modelfusion.dev/integration/model-provider/ollama), [Hugging Face](https://modelfusion.dev/integration/model-provider/huggingface), [Cohere](https://modelfusion.dev/integration/model-provider/cohere)
 
 ### [Tokenize Text](https://modelfusion.dev/guide/function/tokenize-text)
 
@@ -322,7 +289,7 @@ const tokensAndTokenTexts = await tokenizer.tokenizeWithTexts(text);
 const reconstructedText = await tokenizer.detokenize(tokens);
 ```
 
-Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Cohere](https://modelfusion.dev/integration/model-provider/cohere), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp)
+Providers: [OpenAI](https://modelfusion.dev/integration/model-provider/openai), [Llama.cpp](https://modelfusion.dev/integration/model-provider/llamacpp), [Cohere](https://modelfusion.dev/integration/model-provider/cohere)
 
 ### [Guards](https://modelfusion.dev/guide/guard)
 
@@ -366,16 +333,16 @@ Tools are functions that can be executed by an AI model. They are useful for bui
 
 Predefined tools: [SerpAPI](https://modelfusion.dev/integration/tool/serpapi), [Google Custom Search](https://modelfusion.dev/integration/tool/google-custom-search)
 
-#### Create Tool
+#### [Creating Tools](https://modelfusion.dev/guide/tools/create-tools)
 
-A tool is a function with a name, a description, and a schema for the input parameters.
+A tool is comprised of an async execute function, a name, a description, and a schema for the input parameters.
 
 ```ts
 const calculator = new Tool({
   name: "calculator",
   description: "Execute a calculation",
 
-  inputSchema: new ZodSchema(
+  parameters: new ZodSchema(
     z.object({
       a: z.number().describe("The first number."),
       b: z.number().describe("The second number."),
@@ -402,31 +369,76 @@ const calculator = new Tool({
 });
 ```
 
-#### useTool
+#### [generateToolCall](https://modelfusion.dev/guide/tools/generate-tool-call)
 
-The model determines the parameters for the tool from the prompt and then executes it.
+With `generateToolCall`, you can generate a tool call for a specific tool with a language model that supports tools calls (e.g. OpenAI Chat). This function does not execute the tools.
 
 ```ts
-const { tool, parameters, result } = await useTool(
+const { id, name, args } = await generateToolCall(
   new OpenAIChatModel({ model: "gpt-3.5-turbo" }),
   calculator,
   [OpenAIChatMessage.user("What's fourteen times twelve?")]
 );
 ```
 
-#### useToolOrGenerateText
+#### [generateToolCallsOrText](https://modelfusion.dev/guide/tools/generate-tool-calls-or-text)
 
-The model determines which tool to use and its parameters from the prompt and then executes it.
-Text is generated as a fallback.
+With `generateToolCallsOrText`, you can ask a language model to generate several tool calls as well as text. The model will choose which tools (if any) should be called with which arguments. Both the text and the tool calls are optional. This function does not execute the tools.
 
 ```ts
-const { tool, parameters, result, text } = await useToolOrGenerateText(
+const { text, toolCalls } = await generateToolCallsOrText(
   new OpenAIChatModel({ model: "gpt-3.5-turbo" }),
-  [calculator /* and other tools... */],
+  [toolA, toolB, toolC],
+  [OpenAIChatMessage.user(query)]
+);
+```
+
+#### [executeTool](https://modelfusion.dev/guide/tools/execute-tool)
+
+You can directly invoke a tool with `executeTool`:
+
+```ts
+const result = await executeTool(calculator, {
+  a: 14,
+  b: 12,
+  operator: "*",
+});
+```
+
+#### [useTool](https://modelfusion.dev/guide/tools/use-tool)
+
+With `useTool`, you can use a tool with a language model that supports tools calls (e.g. OpenAI Chat). `useTool` first generates a tool call and then executes the tool with the arguments.
+
+```ts
+const { tool, toolCall, args, ok, result } = await useTool(
+  new OpenAIChatModel({ model: "gpt-3.5-turbo" }),
+  calculator,
+  [OpenAIChatMessage.user("What's fourteen times twelve?")]
+);
+
+console.log(`Tool call:`, toolCall);
+console.log(`Tool:`, tool);
+console.log(`Arguments:`, args);
+console.log(`Ok:`, ok);
+console.log(`Result or Error:`, result);
+```
+
+#### [useToolsOrGenerateText](https://modelfusion.dev/guide/tools/use-tools-or-generate-text)
+
+With `useToolsOrGenerateText`, you can ask a language model to generate several tool calls as well as text. The model will choose which tools (if any) should be called with which arguments. Both the text and the tool calls are optional. This function executes the tools.
+
+```ts
+const { text, toolResults } = await useToolsOrGenerateText(
+  new OpenAIChatModel({ model: "gpt-3.5-turbo" }),
+  [calculator /* ... */],
   [OpenAIChatMessage.user("What's fourteen times twelve?")]
 );
 ```
 
+#### [Agent Loop](https://modelfusion.dev/guide/tools/agent-loop)
+
+You can use `useToolsOrGenerateText` to implement an agent loop that responds to user messages and executes tools. [Learn more](https://modelfusion.dev/guide/tools/agent-loop).
+
 ### [Vector Indices](https://modelfusion.dev/guide/vector-index)
 
 ```ts

package/core/FunctionEvent.d.ts

@@ -1,11 +1,11 @@
-import { ExecuteFunctionFinishedEvent, ExecuteFunctionStartedEvent } from "./ExecuteFunctionEvent.js";
 import { GuardFinishedEvent, GuardStartedEvent } from "../guard/GuardEvent.js";
 import { ModelCallFinishedEvent, ModelCallStartedEvent } from "../model-function/ModelCallEvent.js";
 import { RetrieveFinishedEvent, RetrieveStartedEvent } from "../retriever/RetrieveEvent.js";
-import { ExecuteToolFinishedEvent, ExecuteToolStartedEvent } from "../tool/ExecuteToolEvent.js";
-import { UseToolFinishedEvent, UseToolStartedEvent } from "../tool/UseToolEvent.js";
-import { UseToolOrGenerateTextFinishedEvent, UseToolOrGenerateTextStartedEvent } from "../tool/UseToolOrGenerateTextEvent.js";
+import { ExecuteToolFinishedEvent, ExecuteToolStartedEvent } from "../tool/execute-tool/ExecuteToolEvent.js";
+import { UseToolFinishedEvent, UseToolStartedEvent } from "../tool/use-tool/UseToolEvent.js";
+import { UseToolsOrGenerateTextFinishedEvent, UseToolsOrGenerateTextStartedEvent } from "../tool/use-tools-or-generate-text/UseToolsOrGenerateTextEvent.js";
 import { UpsertIntoVectorIndexFinishedEvent, UpsertIntoVectorIndexStartedEvent } from "../vector-index/UpsertIntoVectorIndexEvent.js";
+import { ExecuteFunctionFinishedEvent, ExecuteFunctionStartedEvent } from "./ExecuteFunctionEvent.js";
 export interface BaseFunctionEvent {
     /**
      * Unique identifier for the function call.
@@ -82,4 +82,4 @@ export interface BaseFunctionFinishedEvent extends BaseFunctionEvent {
      */
     result: BaseFunctionFinishedEventResult;
 }
-export type FunctionEvent = ExecuteFunctionStartedEvent | ExecuteFunctionFinishedEvent | ExecuteToolStartedEvent | ExecuteToolFinishedEvent | GuardStartedEvent | GuardFinishedEvent | ModelCallStartedEvent | ModelCallFinishedEvent | RetrieveStartedEvent | RetrieveFinishedEvent | UpsertIntoVectorIndexStartedEvent | UpsertIntoVectorIndexFinishedEvent | UseToolStartedEvent | UseToolFinishedEvent | UseToolOrGenerateTextStartedEvent | UseToolOrGenerateTextFinishedEvent;
+export type FunctionEvent = ExecuteFunctionStartedEvent | ExecuteFunctionFinishedEvent | ExecuteToolStartedEvent | ExecuteToolFinishedEvent | GuardStartedEvent | GuardFinishedEvent | ModelCallStartedEvent | ModelCallFinishedEvent | RetrieveStartedEvent | RetrieveFinishedEvent | UpsertIntoVectorIndexStartedEvent | UpsertIntoVectorIndexFinishedEvent | UseToolStartedEvent | UseToolFinishedEvent | UseToolsOrGenerateTextStartedEvent | UseToolsOrGenerateTextFinishedEvent;

package/guard/guard.cjs

@@ -4,14 +4,14 @@ exports.guard = void 0;
 const executeFunctionCall_js_1 = require("../core/executeFunctionCall.cjs");
 async function guard(execute, input, guards, options) {
     const guardList = Array.isArray(guards) ? guards : [guards];
-    const maxRetries = options?.maxRetries ?? 1;
+    const maxAttempts = options?.maxAttempts ?? 2;
     return (0, executeFunctionCall_js_1.executeFunctionCall)({
         options,
         input,
         functionType: "guard",
         execute: async (options) => {
             let attempts = 0;
-            while (attempts <= maxRetries) {
+            while (attempts < maxAttempts) {
                 let result;
                 try {
                     result = {
@@ -70,8 +70,9 @@ async function guard(execute, input, guards, options) {
                 }
                 attempts++;
             }
-            throw new Error(`Maximum retry attempts of ${maxRetries} reached ` +
-                `without producing a valid output or handling an error after ${attempts} attempts.`);
+            // TODO dedicated error type
+            throw new Error(`Maximum attempts of ${maxAttempts} reached ` +
+                `without producing a valid output or handling an error.`);
         },
     });
 }

package/guard/guard.d.ts

@@ -24,6 +24,6 @@ export type Guard<INPUT, OUTPUT> = ({ type, input, output, error, }: OutputResul
     action: "passThrough";
 } | undefined>;
 export declare function guard<INPUT, OUTPUT>(execute: (input: INPUT, options?: FunctionOptions) => PromiseLike<OUTPUT>, input: INPUT, guards: Guard<INPUT, OUTPUT> | Array<Guard<INPUT, OUTPUT>>, options?: FunctionOptions & {
-    maxRetries: number;
+    maxAttempts: number;
 }): Promise<OUTPUT | undefined>;
 export {};

package/guard/guard.js

@@ -1,14 +1,14 @@
 import { executeFunctionCall } from "../core/executeFunctionCall.js";
 export async function guard(execute, input, guards, options) {
     const guardList = Array.isArray(guards) ? guards : [guards];
-    const maxRetries = options?.maxRetries ?? 1;
+    const maxAttempts = options?.maxAttempts ?? 2;
     return executeFunctionCall({
         options,
         input,
         functionType: "guard",
         execute: async (options) => {
             let attempts = 0;
-            while (attempts <= maxRetries) {
+            while (attempts < maxAttempts) {
                 let result;
                 try {
                     result = {
@@ -67,8 +67,9 @@ export async function guard(execute, input, guards, options) {
                 }
                 attempts++;
             }
-            throw new Error(`Maximum retry attempts of ${maxRetries} reached ` +
-                `without producing a valid output or handling an error after ${attempts} attempts.`);
+            // TODO dedicated error type
+            throw new Error(`Maximum attempts of ${maxAttempts} reached ` +
+                `without producing a valid output or handling an error.`);
         },
     });
 }

package/model-function/ModelCallEvent.d.ts

@@ -1,4 +1,6 @@
 import { BaseFunctionFinishedEvent, BaseFunctionStartedEvent } from "../core/FunctionEvent.js";
+import { ToolCallGenerationFinishedEvent, ToolCallGenerationStartedEvent } from "../tool/generate-tool-call/ToolCallGenerationEvent.js";
+import { ToolCallsOrTextGenerationFinishedEvent, ToolCallsOrTextGenerationStartedEvent } from "../tool/generate-tool-calls-or-text/ToolCallsOrTextGenerationEvent.js";
 import { ModelInformation } from "./ModelInformation.js";
 import { EmbeddingFinishedEvent, EmbeddingStartedEvent } from "./embed/EmbeddingEvent.js";
 import { ImageGenerationFinishedEvent, ImageGenerationStartedEvent } from "./generate-image/ImageGenerationEvent.js";
@@ -6,7 +8,6 @@ import { SpeechGenerationFinishedEvent, SpeechGenerationStartedEvent, SpeechStre
 import { StructureGenerationFinishedEvent, StructureGenerationStartedEvent } from "./generate-structure/StructureGenerationEvent.js";
 import { StructureStreamingFinishedEvent, StructureStreamingStartedEvent } from "./generate-structure/StructureStreamingEvent.js";
 import { TextGenerationFinishedEvent, TextGenerationStartedEvent, TextStreamingFinishedEvent, TextStreamingStartedEvent } from "./generate-text/TextGenerationEvent.js";
-import { ToolCallGenerationFinishedEvent, ToolCallGenerationStartedEvent } from "./generate-tool-call/ToolCallGenerationEvent.js";
 import { TranscriptionFinishedEvent, TranscriptionStartedEvent } from "./generate-transcription/TranscriptionEvent.js";
 export interface BaseModelCallStartedEvent extends BaseFunctionStartedEvent {
     model: ModelInformation;
@@ -48,5 +49,5 @@ export interface BaseModelCallFinishedEvent extends BaseFunctionFinishedEvent {
      */
     result: BaseModelCallFinishedEventResult;
 }
-export type ModelCallStartedEvent = EmbeddingStartedEvent | ImageGenerationStartedEvent | SpeechGenerationStartedEvent | SpeechStreamingStartedEvent | StructureGenerationStartedEvent | StructureStreamingStartedEvent | TextGenerationStartedEvent | TextStreamingStartedEvent | ToolCallGenerationStartedEvent | TranscriptionStartedEvent;
-export type ModelCallFinishedEvent = EmbeddingFinishedEvent | ImageGenerationFinishedEvent | SpeechGenerationFinishedEvent | SpeechStreamingFinishedEvent | StructureGenerationFinishedEvent | StructureStreamingFinishedEvent | TextGenerationFinishedEvent | TextStreamingFinishedEvent | ToolCallGenerationFinishedEvent | TranscriptionFinishedEvent;
+export type ModelCallStartedEvent = EmbeddingStartedEvent | ImageGenerationStartedEvent | SpeechGenerationStartedEvent | SpeechStreamingStartedEvent | StructureGenerationStartedEvent | StructureStreamingStartedEvent | TextGenerationStartedEvent | TextStreamingStartedEvent | ToolCallGenerationStartedEvent | ToolCallsOrTextGenerationStartedEvent | TranscriptionStartedEvent;
+export type ModelCallFinishedEvent = EmbeddingFinishedEvent | ImageGenerationFinishedEvent | SpeechGenerationFinishedEvent | SpeechStreamingFinishedEvent | StructureGenerationFinishedEvent | StructureStreamingFinishedEvent | TextGenerationFinishedEvent | TextStreamingFinishedEvent | ToolCallGenerationFinishedEvent | ToolCallsOrTextGenerationFinishedEvent | TranscriptionFinishedEvent;

package/model-function/generate-structure/StructureGenerationEvent.d.ts

@@ -1,6 +1,6 @@
 import { BaseModelCallFinishedEvent, BaseModelCallStartedEvent } from "../ModelCallEvent.js";
 export interface StructureGenerationStartedEvent extends BaseModelCallStartedEvent {
-    functionType: "generate-structure" | "generate-structure-or-text";
+    functionType: "generate-structure";
 }
 export type StructureGenerationFinishedEventResult = {
     status: "success";
@@ -18,6 +18,6 @@ export type StructureGenerationFinishedEventResult = {
     status: "abort";
 };
 export type StructureGenerationFinishedEvent = BaseModelCallFinishedEvent & {
-    functionType: "generate-structure" | "generate-structure-or-text";
+    functionType: "generate-structure";
     result: StructureGenerationFinishedEventResult;
 };

package/model-function/generate-structure/StructureParseError.cjs

@@ -30,5 +30,15 @@ class StructureParseError extends Error {
         this.cause = cause;
         this.valueText = valueText;
     }
+    toJSON() {
+        return {
+            name: this.name,
+            cause: this.cause,
+            message: this.message,
+            stack: this.stack,
+            structureName: this.structureName,
+            valueText: this.valueText,
+        };
+    }
 }
 exports.StructureParseError = StructureParseError;

package/model-function/generate-structure/StructureParseError.d.ts

@@ -7,4 +7,12 @@ export declare class StructureParseError extends Error {
         valueText: string;
         cause: unknown;
     });
+    toJSON(): {
+        name: string;
+        cause: unknown;
+        message: string;
+        stack: string | undefined;
+        structureName: string;
+        valueText: string;
+    };
 }

package/model-function/generate-structure/StructureParseError.js

@@ -27,4 +27,14 @@ export class StructureParseError extends Error {
         this.cause = cause;
         this.valueText = valueText;
     }
+    toJSON() {
+        return {
+            name: this.name,
+            cause: this.cause,
+            message: this.message,
+            stack: this.stack,
+            structureName: this.structureName,
+            valueText: this.valueText,
+        };
+    }
 }

package/model-function/generate-structure/index.cjs

@@ -14,14 +14,11 @@ var __exportStar = (this && this.__exportStar) || function(m, exports) {
     for (var p in m) if (p !== "default" && !Object.prototype.hasOwnProperty.call(exports, p)) __createBinding(exports, m, p);
 };
 Object.defineProperty(exports, "__esModule", { value: true });
-__exportStar(require("./NoSuchStructureError.cjs"), exports);
 __exportStar(require("./StructureFromTextGenerationModel.cjs"), exports);
 __exportStar(require("./StructureGenerationEvent.cjs"), exports);
 __exportStar(require("./StructureGenerationModel.cjs"), exports);
-__exportStar(require("./StructureOrTextGenerationModel.cjs"), exports);
 __exportStar(require("./StructureParseError.cjs"), exports);
 __exportStar(require("./StructureStreamingEvent.cjs"), exports);
 __exportStar(require("./StructureValidationError.cjs"), exports);
 __exportStar(require("./generateStructure.cjs"), exports);
-__exportStar(require("./generateStructureOrText.cjs"), exports);
 __exportStar(require("./streamStructure.cjs"), exports);

package/model-function/generate-structure/index.d.ts

@@ -1,11 +1,8 @@
-export * from "./NoSuchStructureError.js";
 export * from "./StructureFromTextGenerationModel.js";
 export * from "./StructureGenerationEvent.js";
 export * from "./StructureGenerationModel.js";
-export * from "./StructureOrTextGenerationModel.js";
 export * from "./StructureParseError.js";
 export * from "./StructureStreamingEvent.js";
 export * from "./StructureValidationError.js";
 export * from "./generateStructure.js";
-export * from "./generateStructureOrText.js";
 export * from "./streamStructure.js";

package/model-function/generate-structure/index.js

@@ -1,11 +1,8 @@
-export * from "./NoSuchStructureError.js";
 export * from "./StructureFromTextGenerationModel.js";
 export * from "./StructureGenerationEvent.js";
 export * from "./StructureGenerationModel.js";
-export * from "./StructureOrTextGenerationModel.js";
 export * from "./StructureParseError.js";
 export * from "./StructureStreamingEvent.js";
 export * from "./StructureValidationError.js";
 export * from "./generateStructure.js";
-export * from "./generateStructureOrText.js";
 export * from "./streamStructure.js";

package/model-function/index.cjs

@@ -31,7 +31,6 @@ __exportStar(require("./generate-image/generateImage.cjs"), exports);
 __exportStar(require("./generate-speech/index.cjs"), exports);
 __exportStar(require("./generate-structure/index.cjs"), exports);
 __exportStar(require("./generate-text/index.cjs"), exports);
-__exportStar(require("./generate-tool-call/index.cjs"), exports);
 __exportStar(require("./generate-transcription/TranscriptionEvent.cjs"), exports);
 __exportStar(require("./generate-transcription/TranscriptionModel.cjs"), exports);
 __exportStar(require("./generate-transcription/generateTranscription.cjs"), exports);

package/model-function/index.d.ts

@@ -15,7 +15,6 @@ export * from "./generate-image/generateImage.js";
 export * from "./generate-speech/index.js";
 export * from "./generate-structure/index.js";
 export * from "./generate-text/index.js";
-export * from "./generate-tool-call/index.js";
 export * from "./generate-transcription/TranscriptionEvent.js";
 export * from "./generate-transcription/TranscriptionModel.js";
 export * from "./generate-transcription/generateTranscription.js";

package/model-function/index.js

@@ -15,7 +15,6 @@ export * from "./generate-image/generateImage.js";
 export * from "./generate-speech/index.js";
 export * from "./generate-structure/index.js";
 export * from "./generate-text/index.js";
-export * from "./generate-tool-call/index.js";
 export * from "./generate-transcription/TranscriptionEvent.js";
 export * from "./generate-transcription/TranscriptionModel.js";
 export * from "./generate-transcription/generateTranscription.js";

package/model-provider/automatic1111/Automatic1111ImageGenerationModel.d.ts

@@ -19,8 +19,8 @@ export declare class Automatic1111ImageGenerationModel extends AbstractModel<Aut
     get settingsForEvent(): Partial<Automatic1111ImageGenerationSettings>;
     doGenerateImage(prompt: Automatic1111ImageGenerationPrompt, options?: FunctionOptions): Promise<{
         response: {
-            parameters: {};
             images: string[];
+            parameters: {};
             info: string;
         };
         base64Image: string;
@@ -42,12 +42,12 @@ declare const Automatic1111ImageGenerationResponseSchema: z.ZodObject<{
     parameters: z.ZodObject<{}, "strip", z.ZodTypeAny, {}, {}>;
     info: z.ZodString;
 }, "strip", z.ZodTypeAny, {
-    parameters: {};
     images: string[];
+    parameters: {};
     info: string;
 }, {
-    parameters: {};
     images: string[];
+    parameters: {};
     info: string;
 }>;
 export type Automatic1111ImageGenerationResponse = z.infer<typeof Automatic1111ImageGenerationResponseSchema>;

package/model-provider/openai/chat/OpenAIChatMessage.cjs

@@ -21,40 +21,31 @@ exports.OpenAIChatMessage = {
         }
         return { role: "user", content, name: options?.name };
     },
-    assistant(content) {
-        return { role: "assistant", content };
-    },
-    functionCall(content, functionCall) {
-        return {
-            role: "assistant",
-            content,
-            function_call: functionCall,
-        };
-    },
-    functionResult(name, content) {
-        return { role: "function", name, content };
-    },
     /**
-     * Creates a function call chat message for tool calls.
+     * Creates an assistant chat message. The assistant message can optionally contain tool calls.
      */
-    toolCall({ text, tool, parameters, }) {
+    assistant(content, options) {
         return {
             role: "assistant",
-            content: text,
-            function_call: {
-                name: tool,
-                arguments: JSON.stringify(parameters),
-            },
+            content,
+            tool_calls: options?.toolCalls?.map((toolCall) => ({
+                id: toolCall.id,
+                type: "function",
+                function: {
+                    name: toolCall.name,
+                    arguments: JSON.stringify(toolCall.args),
+                },
+            })) ?? undefined,
         };
     },
     /**
-     * Creates a function result chat message for tool call results.
+     * Creates a tool result chat message with the result of a tool call.
      */
-    toolResult({ tool, result }) {
+    tool({ toolCallId, content, }) {
         return {
-            role: "function",
-            name: tool,
-            content: JSON.stringify(result),
+            role: "tool",
+            tool_call_id: toolCallId,
+            content: JSON.stringify(content),
         };
     },
 };

package/model-provider/openai/chat/OpenAIChatMessage.d.ts

@@ -1,3 +1,4 @@
+import { ToolCall } from "../../../tool/ToolCall.js";
 export type OpenAIChatMessage = {
     role: "system";
     content: string;
@@ -9,7 +10,10 @@ export type OpenAIChatMessage = {
         text: string;
     } | {
         type: "image_url";
-        image_url: string;
+        image_url: string | {
+            url: string;
+            detail: "low" | "high" | "auto";
+        };
     }>;
     name?: string;
 } | {
@@ -46,36 +50,17 @@ export declare const OpenAIChatMessage: {
             mimeType?: string;
         };
     }): OpenAIChatMessage;
-    assistant(content: string): OpenAIChatMessage;
-    functionCall(content: string | null, functionCall: {
-        name: string;
-        arguments: string;
-    }): OpenAIChatMessage;
-    functionResult(name: string, content: string): OpenAIChatMessage;
     /**
-     * Creates a function call chat message for tool calls.
+     * Creates an assistant chat message. The assistant message can optionally contain tool calls.
      */
-    toolCall({ text, tool, parameters, }: {
-        text: string | null;
-        tool: string;
-        parameters: unknown;
-    }): {
-        role: "assistant";
-        content: string | null;
-        function_call: {
-            name: string;
-            arguments: string;
-        };
-    };
+    assistant(content: string | null, options?: {
+        toolCalls: Array<ToolCall<string, unknown>> | null | undefined;
+    }): OpenAIChatMessage;
     /**
-     * Creates a function result chat message for tool call results.
+     * Creates a tool result chat message with the result of a tool call.
      */
-    toolResult({ tool, result }: {
-        tool: string;
-        result: unknown;
-    }): {
-        role: "function";
-        name: string;
-        content: string;
-    };
+    tool({ toolCallId, content, }: {
+        toolCallId: string;
+        content: unknown;
+    }): OpenAIChatMessage;
 };

package/model-provider/openai/chat/OpenAIChatMessage.js

@@ -18,40 +18,31 @@ export const OpenAIChatMessage = {
         }
         return { role: "user", content, name: options?.name };
     },
-    assistant(content) {
-        return { role: "assistant", content };
-    },
-    functionCall(content, functionCall) {
-        return {
-            role: "assistant",
-            content,
-            function_call: functionCall,
-        };
-    },
-    functionResult(name, content) {
-        return { role: "function", name, content };
-    },
     /**
-     * Creates a function call chat message for tool calls.
+     * Creates an assistant chat message. The assistant message can optionally contain tool calls.
      */
-    toolCall({ text, tool, parameters, }) {
+    assistant(content, options) {
         return {
             role: "assistant",
-            content: text,
-            function_call: {
-                name: tool,
-                arguments: JSON.stringify(parameters),
-            },
+            content,
+            tool_calls: options?.toolCalls?.map((toolCall) => ({
+                id: toolCall.id,
+                type: "function",
+                function: {
+                    name: toolCall.name,
+                    arguments: JSON.stringify(toolCall.args),
+                },
+            })) ?? undefined,
         };
     },
     /**
-     * Creates a function result chat message for tool call results.
+     * Creates a tool result chat message with the result of a tool call.
      */
-    toolResult({ tool, result }) {
+    tool({ toolCallId, content, }) {
         return {
-            role: "function",
-            name: tool,
-            content: JSON.stringify(result),
+            role: "tool",
+            tool_call_id: toolCallId,
+            content: JSON.stringify(content),
         };
     },
 };