modelfusion 0.54.0 → 0.55.1

package/README.md

@@ -20,6 +20,7 @@
 - **Type inference and validation**: ModelFusion infers TypeScript types wherever possible and to validates model responses.
 - **Observability and logging**: ModelFusion provides an observer framework and out-of-the-box logging support.
 - **Resilience and Robustness**: ModelFusion ensures seamless operation through automatic retries, throttling, and error handling mechanisms.
+- **Server**: ModelFusion provides a Fastify plugin that exposes a ModelFusion flow as a REST endpoint that uses server-sent events.
 
 ## Quick Install
 
@@ -53,7 +54,7 @@ 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), [Hugging Face](https://modelfusion.dev/integration/model-provider/huggingface)
+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)
 
 #### streamText
 
@@ -70,7 +71,7 @@ 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)
+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)
 
 ### [Generate Image](https://modelfusion.dev/guide/function/generate-image)
 
@@ -543,6 +544,71 @@ ModelFusion provides an [observer framework](https://modelfusion.dev/guide/util/
 setGlobalFunctionLogging("detailed-object"); // log full events
 ```
 
+### [Server](https://modelfusion.dev/guide/server/)
+
+> [!WARNING]
+> ModelFusion Server is in its initial development phase and not feature-complete. The API is experimental and breaking changes are likely. Feedback and suggestions are welcome.
+
+ModelFusion Server is desigend for running multi-modal generative AI flows that take up to several minutes to complete. It provides the following benefits:
+
+- 🔄 Real-time progress updates via custom server-sent events
+- 🔒Type-safety with Zod-schema for inputs/events
+- 📦 Efficient handling of dynamically created binary assets (images, audio)
+- 📜 Auto-logging for AI model interactions within flows
+
+ModelFusion provides a [Fastify](https://fastify.dev/) plugin that allows you to set up a server that exposes your ModelFusion flows as REST endpoints using server-sent events.
+
+```ts
+import {
+  FileSystemAssetStorage,
+  FileSystemLogger,
+  modelFusionFastifyPlugin,
+} from "modelfusion/fastify-server"; // '/fastify-server' import path
+
+// configurable logging for all runs using ModelFusion observability:
+const logger = new FileSystemLogger({
+  path: (run) => path.join(fsBasePath, run.runId, "logs"),
+});
+
+// configurable storage for large files like images and audio files:
+const assetStorage = new FileSystemAssetStorage({
+  path: (run) => path.join(fsBasePath, run.runId, "assets"),
+  logger,
+});
+
+fastify.register(modelFusionFastifyPlugin, {
+  baseUrl,
+  basePath: "/myFlow",
+  logger,
+  assetStorage,
+  flow: exampleFlow,
+});
+```
+
+Using `invokeFlow`, you can easily connect your client to a ModelFusion flow endpoint:
+
+```ts
+import { invokeFlow } from "modelfusion/browser"; // '/browser' import path
+
+invokeFlow({
+  url: `${BASE_URL}/myFlow`,
+  schema: myFlowSchema,
+  input: { prompt },
+  onEvent(event) {
+    switch (event.type) {
+      case "my-event": {
+        // do something with the event
+        break;
+      }
+      // more events...
+    }
+  },
+  onStop() {
+    // flow finished
+  },
+});
+```
+
 ## Documentation
 
 ### [Guide](https://modelfusion.dev/guide)
@@ -563,6 +629,7 @@ setGlobalFunctionLogging("detailed-object"); // log full events
   - [Retrieve](https://modelfusion.dev/guide/vector-index/retrieve)
 - [Text Chunks](https://modelfusion.dev/guide/text-chunk/)
   - [Split Text](https://modelfusion.dev/guide/text-chunk/split)
+- [Server](https://modelfusion.dev/guide/server/)
 - [Utilities](https://modelfusion.dev/guide/util/)
   - [API Configuration](https://modelfusion.dev/guide/util/api-configuration)
     - [Retry strategies](https://modelfusion.dev/guide/util/api-configuration/retry)

package/browser/readEventSourceStream.cjs

@@ -3,7 +3,7 @@ Object.defineProperty(exports, "__esModule", { value: true });
 exports.readEventSourceStream = void 0;
 const parseJSON_js_1 = require("../util/parseJSON.cjs");
 const AsyncQueue_js_1 = require("../util/AsyncQueue.cjs");
-const parseEventSourceStream_js_1 = require("../event-source/parseEventSourceStream.cjs");
+const parseEventSourceStream_js_1 = require("../util/streaming/parseEventSourceStream.cjs");
 function readEventSourceStream({ stream, schema, errorHandler, }) {
     const queue = new AsyncQueue_js_1.AsyncQueue();
     // run async (no await on purpose):

package/browser/readEventSourceStream.js

@@ -1,6 +1,6 @@
 import { safeParseJsonWithSchema } from "../util/parseJSON.js";
 import { AsyncQueue } from "../util/AsyncQueue.js";
-import { parseEventSourceStream } from "../event-source/parseEventSourceStream.js";
+import { parseEventSourceStream } from "../util/streaming/parseEventSourceStream.js";
 export function readEventSourceStream({ stream, schema, errorHandler, }) {
     const queue = new AsyncQueue();
     // run async (no await on purpose):

package/index.cjs

@@ -17,7 +17,6 @@ Object.defineProperty(exports, "__esModule", { value: true });
 __exportStar(require("./composed-function/index.cjs"), exports);
 __exportStar(require("./core/index.cjs"), exports);
 __exportStar(require("./cost/index.cjs"), exports);
-__exportStar(require("./event-source/index.cjs"), exports);
 __exportStar(require("./guard/index.cjs"), exports);
 __exportStar(require("./model-function/index.cjs"), exports);
 __exportStar(require("./model-provider/index.cjs"), exports);

package/index.d.ts

@@ -1,7 +1,6 @@
 export * from "./composed-function/index.js";
 export * from "./core/index.js";
 export * from "./cost/index.js";
-export * from "./event-source/index.js";
 export * from "./guard/index.js";
 export * from "./model-function/index.js";
 export * from "./model-provider/index.js";

package/index.js

@@ -1,7 +1,6 @@
 export * from "./composed-function/index.js";
 export * from "./core/index.js";
 export * from "./cost/index.js";
-export * from "./event-source/index.js";
 export * from "./guard/index.js";
 export * from "./model-function/index.js";
 export * from "./model-provider/index.js";

package/model-function/embed/EmbeddingModel.d.ts

@@ -12,6 +12,10 @@ export interface EmbeddingModel<VALUE, SETTINGS extends EmbeddingModelSettings =
      * Limit of how many values can be sent in a single API call.
      */
     readonly maxValuesPerCall: number | undefined;
+    /**
+     * True if the model can handle multiple embedding calls in parallel.
+     */
+    readonly isParallizable: boolean;
     doEmbedValues(values: VALUE[], options?: FunctionOptions): PromiseLike<{
         response: unknown;
         embeddings: Vector[];

package/model-function/embed/embed.cjs

@@ -41,7 +41,18 @@ function embedMany(model, values, options) {
                     valueGroups.push(values.slice(i, i + maxValuesPerCall));
                 }
             }
-            const responses = await Promise.all(valueGroups.map((valueGroup) => model.doEmbedValues(valueGroup, options)));
+            // call the model for each group:
+            let responses;
+            if (model.isParallizable) {
+                responses = await Promise.all(valueGroups.map((valueGroup) => model.doEmbedValues(valueGroup, options)));
+            }
+            else {
+                responses = [];
+                for (const valueGroup of valueGroups) {
+                    const response = await model.doEmbedValues(valueGroup, options);
+                    responses.push(response);
+                }
+            }
             const rawResponses = responses.map((response) => response.response);
             const embeddings = [];
             for (const response of responses) {

package/model-function/embed/embed.js

@@ -38,7 +38,18 @@ export function embedMany(model, values, options) {
                     valueGroups.push(values.slice(i, i + maxValuesPerCall));
                 }
             }
-            const responses = await Promise.all(valueGroups.map((valueGroup) => model.doEmbedValues(valueGroup, options)));
+            // call the model for each group:
+            let responses;
+            if (model.isParallizable) {
+                responses = await Promise.all(valueGroups.map((valueGroup) => model.doEmbedValues(valueGroup, options)));
+            }
+            else {
+                responses = [];
+                for (const valueGroup of valueGroups) {
+                    const response = await model.doEmbedValues(valueGroup, options);
+                    responses.push(response);
+                }
+            }
             const rawResponses = responses.map((response) => response.response);
             const embeddings = [];
             for (const response of responses) {

package/model-provider/anthropic/AnthropicTextGenerationModel.cjs

@@ -4,7 +4,7 @@ exports.AnthropicTextGenerationResponseFormat = exports.AnthropicTextGenerationM
 const zod_1 = require("zod");
 const callWithRetryAndThrottle_js_1 = require("../../core/api/callWithRetryAndThrottle.cjs");
 const postToApi_js_1 = require("../../core/api/postToApi.cjs");
-const parseEventSourceStream_js_1 = require("../../event-source/parseEventSourceStream.cjs");
+const parseEventSourceStream_js_1 = require("../../util/streaming/parseEventSourceStream.cjs");
 const AbstractModel_js_1 = require("../../model-function/AbstractModel.cjs");
 const PromptFormatTextStreamingModel_js_1 = require("../../model-function/generate-text/PromptFormatTextStreamingModel.cjs");
 const AsyncQueue_js_1 = require("../../util/AsyncQueue.cjs");

package/model-provider/anthropic/AnthropicTextGenerationModel.js

@@ -1,7 +1,7 @@
 import { z } from "zod";
 import { callWithRetryAndThrottle } from "../../core/api/callWithRetryAndThrottle.js";
 import { createJsonResponseHandler, postJsonToApi, } from "../../core/api/postToApi.js";
-import { parseEventSourceStream } from "../../event-source/parseEventSourceStream.js";
+import { parseEventSourceStream } from "../../util/streaming/parseEventSourceStream.js";
 import { AbstractModel } from "../../model-function/AbstractModel.js";
 import { PromptFormatTextStreamingModel } from "../../model-function/generate-text/PromptFormatTextStreamingModel.js";
 import { AsyncQueue } from "../../util/AsyncQueue.js";

package/model-provider/cohere/CohereTextEmbeddingModel.cjs

@@ -51,6 +51,12 @@ class CohereTextEmbeddingModel extends AbstractModel_js_1.AbstractModel {
             writable: true,
             value: 96
         });
+        Object.defineProperty(this, "isParallizable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
         Object.defineProperty(this, "embeddingDimensions", {
             enumerable: true,
             configurable: true,

package/model-provider/cohere/CohereTextEmbeddingModel.d.ts

@@ -43,6 +43,7 @@ export declare class CohereTextEmbeddingModel extends AbstractModel<CohereTextEm
     readonly provider: "cohere";
     get modelName(): "embed-english-light-v2.0" | "embed-english-v2.0" | "embed-multilingual-v2.0";
     readonly maxValuesPerCall = 96;
+    readonly isParallizable = true;
     readonly embeddingDimensions: number;
     readonly contextWindowSize: number;
     private readonly tokenizer;

package/model-provider/cohere/CohereTextEmbeddingModel.js

@@ -48,6 +48,12 @@ export class CohereTextEmbeddingModel extends AbstractModel {
             writable: true,
             value: 96
         });
+        Object.defineProperty(this, "isParallizable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
         Object.defineProperty(this, "embeddingDimensions", {
             enumerable: true,
             configurable: true,

package/model-provider/cohere/CohereTextGenerationModel.cjs

@@ -4,12 +4,12 @@ exports.CohereTextGenerationResponseFormat = exports.CohereTextGenerationModel =
 const zod_1 = require("zod");
 const callWithRetryAndThrottle_js_1 = require("../../core/api/callWithRetryAndThrottle.cjs");
 const postToApi_js_1 = require("../../core/api/postToApi.cjs");
-const AsyncQueue_js_1 = require("../../util/AsyncQueue.cjs");
 const AbstractModel_js_1 = require("../../model-function/AbstractModel.cjs");
 const PromptFormatTextStreamingModel_js_1 = require("../../model-function/generate-text/PromptFormatTextStreamingModel.cjs");
 const TextPromptFormat_js_1 = require("../../model-function/generate-text/TextPromptFormat.cjs");
 const countTokens_js_1 = require("../../model-function/tokenize-text/countTokens.cjs");
-const parseJSON_js_1 = require("../../util/parseJSON.cjs");
+const AsyncQueue_js_1 = require("../../util/AsyncQueue.cjs");
+const parseJsonStream_js_1 = require("../../util/streaming/parseJsonStream.cjs");
 const CohereApiConfiguration_js_1 = require("./CohereApiConfiguration.cjs");
 const CohereError_js_1 = require("./CohereError.cjs");
 const CohereTokenizer_js_1 = require("./CohereTokenizer.cjs");
@@ -216,58 +216,39 @@ const cohereTextStreamingResponseSchema = zod_1.z.discriminatedUnion("is_finishe
 async function createCohereTextGenerationFullDeltaIterableQueue(stream) {
     const queue = new AsyncQueue_js_1.AsyncQueue();
     let accumulatedText = "";
-    function processLine(line) {
-        const event = (0, parseJSON_js_1.parseJsonWithZod)(line, cohereTextStreamingResponseSchema);
-        if (event.is_finished === true) {
-            queue.push({
-                type: "delta",
-                fullDelta: {
-                    content: accumulatedText,
-                    isComplete: true,
-                    delta: "",
-                },
-                valueDelta: "",
-            });
-        }
-        else {
-            accumulatedText += event.text;
-            queue.push({
-                type: "delta",
-                fullDelta: {
-                    content: accumulatedText,
-                    isComplete: false,
-                    delta: event.text,
-                },
-                valueDelta: event.text,
-            });
-        }
-    }
     // process the stream asynchonously (no 'await' on purpose):
-    (async () => {
-        try {
-            let unprocessedText = "";
-            const reader = new ReadableStreamDefaultReader(stream);
-            const utf8Decoder = new TextDecoder("utf-8");
-            // eslint-disable-next-line no-constant-condition
-            while (true) {
-                const { value: chunk, done } = await reader.read();
-                if (done) {
-                    break;
-                }
-                unprocessedText += utf8Decoder.decode(chunk, { stream: true });
-                const processableLines = unprocessedText.split(/\r\n|\n|\r/g);
-                unprocessedText = processableLines.pop() || "";
-                processableLines.forEach(processLine);
+    (0, parseJsonStream_js_1.parseJsonStream)({
+        stream,
+        schema: cohereTextStreamingResponseSchema,
+        process(event) {
+            if (event.is_finished === true) {
+                queue.push({
+                    type: "delta",
+                    fullDelta: {
+                        content: accumulatedText,
+                        isComplete: true,
+                        delta: "",
+                    },
+                    valueDelta: "",
+                });
             }
-            // processing remaining text:
-            if (unprocessedText) {
-                processLine(unprocessedText);
+            else {
+                accumulatedText += event.text;
+                queue.push({
+                    type: "delta",
+                    fullDelta: {
+                        content: accumulatedText,
+                        isComplete: false,
+                        delta: event.text,
+                    },
+                    valueDelta: event.text,
+                });
             }
-        }
-        finally {
+        },
+        onDone() {
             queue.close();
-        }
-    })();
+        },
+    });
     return queue;
 }
 exports.CohereTextGenerationResponseFormat = {

package/model-provider/cohere/CohereTextGenerationModel.js

@@ -1,12 +1,12 @@
 import { z } from "zod";
 import { callWithRetryAndThrottle } from "../../core/api/callWithRetryAndThrottle.js";
 import { createJsonResponseHandler, postJsonToApi, } from "../../core/api/postToApi.js";
-import { AsyncQueue } from "../../util/AsyncQueue.js";
 import { AbstractModel } from "../../model-function/AbstractModel.js";
 import { PromptFormatTextStreamingModel } from "../../model-function/generate-text/PromptFormatTextStreamingModel.js";
 import { mapChatPromptToTextFormat, mapInstructionPromptToTextFormat, } from "../../model-function/generate-text/TextPromptFormat.js";
 import { countTokens } from "../../model-function/tokenize-text/countTokens.js";
-import { parseJsonWithZod } from "../../util/parseJSON.js";
+import { AsyncQueue } from "../../util/AsyncQueue.js";
+import { parseJsonStream } from "../../util/streaming/parseJsonStream.js";
 import { CohereApiConfiguration } from "./CohereApiConfiguration.js";
 import { failedCohereCallResponseHandler } from "./CohereError.js";
 import { CohereTokenizer } from "./CohereTokenizer.js";
@@ -212,58 +212,39 @@ const cohereTextStreamingResponseSchema = z.discriminatedUnion("is_finished", [
 async function createCohereTextGenerationFullDeltaIterableQueue(stream) {
     const queue = new AsyncQueue();
     let accumulatedText = "";
-    function processLine(line) {
-        const event = parseJsonWithZod(line, cohereTextStreamingResponseSchema);
-        if (event.is_finished === true) {
-            queue.push({
-                type: "delta",
-                fullDelta: {
-                    content: accumulatedText,
-                    isComplete: true,
-                    delta: "",
-                },
-                valueDelta: "",
-            });
-        }
-        else {
-            accumulatedText += event.text;
-            queue.push({
-                type: "delta",
-                fullDelta: {
-                    content: accumulatedText,
-                    isComplete: false,
-                    delta: event.text,
-                },
-                valueDelta: event.text,
-            });
-        }
-    }
     // process the stream asynchonously (no 'await' on purpose):
-    (async () => {
-        try {
-            let unprocessedText = "";
-            const reader = new ReadableStreamDefaultReader(stream);
-            const utf8Decoder = new TextDecoder("utf-8");
-            // eslint-disable-next-line no-constant-condition
-            while (true) {
-                const { value: chunk, done } = await reader.read();
-                if (done) {
-                    break;
-                }
-                unprocessedText += utf8Decoder.decode(chunk, { stream: true });
-                const processableLines = unprocessedText.split(/\r\n|\n|\r/g);
-                unprocessedText = processableLines.pop() || "";
-                processableLines.forEach(processLine);
+    parseJsonStream({
+        stream,
+        schema: cohereTextStreamingResponseSchema,
+        process(event) {
+            if (event.is_finished === true) {
+                queue.push({
+                    type: "delta",
+                    fullDelta: {
+                        content: accumulatedText,
+                        isComplete: true,
+                        delta: "",
+                    },
+                    valueDelta: "",
+                });
             }
-            // processing remaining text:
-            if (unprocessedText) {
-                processLine(unprocessedText);
+            else {
+                accumulatedText += event.text;
+                queue.push({
+                    type: "delta",
+                    fullDelta: {
+                        content: accumulatedText,
+                        isComplete: false,
+                        delta: event.text,
+                    },
+                    valueDelta: event.text,
+                });
             }
-        }
-        finally {
+        },
+        onDone() {
             queue.close();
-        }
-    })();
+        },
+    });
     return queue;
 }
 export const CohereTextGenerationResponseFormat = {

package/model-provider/huggingface/HuggingFaceTextEmbeddingModel.cjs

@@ -42,6 +42,12 @@ class HuggingFaceTextEmbeddingModel extends AbstractModel_js_1.AbstractModel {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "isParallizable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
         Object.defineProperty(this, "contextWindowSize", {
             enumerable: true,
             configurable: true,

package/model-provider/huggingface/HuggingFaceTextEmbeddingModel.d.ts

@@ -38,6 +38,7 @@ export declare class HuggingFaceTextEmbeddingModel extends AbstractModel<Hugging
     readonly provider = "huggingface";
     get modelName(): string;
     readonly maxValuesPerCall: number;
+    readonly isParallizable = true;
     readonly contextWindowSize: undefined;
     readonly embeddingDimensions: number | undefined;
     readonly tokenizer: undefined;

package/model-provider/huggingface/HuggingFaceTextEmbeddingModel.js

@@ -39,6 +39,12 @@ export class HuggingFaceTextEmbeddingModel extends AbstractModel {
             writable: true,
             value: void 0
         });
+        Object.defineProperty(this, "isParallizable", {
+            enumerable: true,
+            configurable: true,
+            writable: true,
+            value: true
+        });
         Object.defineProperty(this, "contextWindowSize", {
             enumerable: true,
             configurable: true,

package/model-provider/index.cjs

@@ -21,5 +21,6 @@ __exportStar(require("./elevenlabs/index.cjs"), exports);
 __exportStar(require("./huggingface/index.cjs"), exports);
 __exportStar(require("./llamacpp/index.cjs"), exports);
 __exportStar(require("./lmnt/index.cjs"), exports);
+__exportStar(require("./ollama/index.cjs"), exports);
 __exportStar(require("./openai/index.cjs"), exports);
 __exportStar(require("./stability/index.cjs"), exports);

package/model-provider/index.d.ts

@@ -5,5 +5,6 @@ export * from "./elevenlabs/index.js";
 export * from "./huggingface/index.js";
 export * from "./llamacpp/index.js";
 export * from "./lmnt/index.js";
+export * from "./ollama/index.js";
 export * from "./openai/index.js";
 export * from "./stability/index.js";

package/model-provider/index.js

@@ -5,5 +5,6 @@ export * from "./elevenlabs/index.js";
 export * from "./huggingface/index.js";
 export * from "./llamacpp/index.js";
 export * from "./lmnt/index.js";
+export * from "./ollama/index.js";
 export * from "./openai/index.js";
 export * from "./stability/index.js";

package/model-provider/llamacpp/LlamaCppTextEmbeddingModel.cjs

@@ -47,6 +47,9 @@ class LlamaCppTextEmbeddingModel extends AbstractModel_js_1.AbstractModel {
     get modelName() {
         return null;
     }
+    get isParallizable() {
+        return this.settings.isParallizable ?? false;
+    }
     async tokenize(text) {
         return this.tokenizer.tokenize(text);
     }

package/model-provider/llamacpp/LlamaCppTextEmbeddingModel.d.ts

@@ -6,12 +6,14 @@ import { EmbeddingModel, EmbeddingModelSettings } from "../../model-function/emb
 export interface LlamaCppTextEmbeddingModelSettings extends EmbeddingModelSettings {
     api?: ApiConfiguration;
     embeddingDimensions?: number;
+    isParallizable?: boolean;
 }
 export declare class LlamaCppTextEmbeddingModel extends AbstractModel<LlamaCppTextEmbeddingModelSettings> implements EmbeddingModel<string, LlamaCppTextEmbeddingModelSettings> {
     constructor(settings?: LlamaCppTextEmbeddingModelSettings);
     readonly provider: "llamacpp";
     get modelName(): null;
     readonly maxValuesPerCall = 1;
+    get isParallizable(): boolean;
     readonly contextWindowSize: undefined;
     readonly embeddingDimensions: number | undefined;
     private readonly tokenizer;

package/model-provider/llamacpp/LlamaCppTextEmbeddingModel.js

@@ -44,6 +44,9 @@ export class LlamaCppTextEmbeddingModel extends AbstractModel {
     get modelName() {
         return null;
     }
+    get isParallizable() {
+        return this.settings.isParallizable ?? false;
+    }
     async tokenize(text) {
         return this.tokenizer.tokenize(text);
     }

package/model-provider/llamacpp/LlamaCppTextGenerationModel.cjs

@@ -5,7 +5,7 @@ const zod_1 = require("zod");
 const callWithRetryAndThrottle_js_1 = require("../../core/api/callWithRetryAndThrottle.cjs");
 const postToApi_js_1 = require("../../core/api/postToApi.cjs");
 const AsyncQueue_js_1 = require("../../util/AsyncQueue.cjs");
-const parseEventSourceStream_js_1 = require("../../event-source/parseEventSourceStream.cjs");
+const parseEventSourceStream_js_1 = require("../../util/streaming/parseEventSourceStream.cjs");
 const AbstractModel_js_1 = require("../../model-function/AbstractModel.cjs");
 const PromptFormatTextStreamingModel_js_1 = require("../../model-function/generate-text/PromptFormatTextStreamingModel.cjs");
 const parseJSON_js_1 = require("../../util/parseJSON.cjs");

package/model-provider/llamacpp/LlamaCppTextGenerationModel.js

@@ -2,7 +2,7 @@ import { z } from "zod";
 import { callWithRetryAndThrottle } from "../../core/api/callWithRetryAndThrottle.js";
 import { createJsonResponseHandler, postJsonToApi, } from "../../core/api/postToApi.js";
 import { AsyncQueue } from "../../util/AsyncQueue.js";
-import { parseEventSourceStream } from "../../event-source/parseEventSourceStream.js";
+import { parseEventSourceStream } from "../../util/streaming/parseEventSourceStream.js";
 import { AbstractModel } from "../../model-function/AbstractModel.js";
 import { PromptFormatTextStreamingModel } from "../../model-function/generate-text/PromptFormatTextStreamingModel.js";
 import { parseJsonWithZod } from "../../util/parseJSON.js";

package/model-provider/ollama/OllamaApiConfiguration.cjs

@@ -0,0 +1,15 @@
+"use strict";
+Object.defineProperty(exports, "__esModule", { value: true });
+exports.OllamaApiConfiguration = void 0;
+const BaseUrlApiConfiguration_js_1 = require("../../core/api/BaseUrlApiConfiguration.cjs");
+class OllamaApiConfiguration extends BaseUrlApiConfiguration_js_1.BaseUrlApiConfiguration {
+    constructor({ baseUrl = "http://127.0.0.1:11434", retry, throttle, } = {}) {
+        super({
+            baseUrl,
+            headers: {},
+            retry,
+            throttle,
+        });
+    }
+}
+exports.OllamaApiConfiguration = OllamaApiConfiguration;

package/model-provider/ollama/OllamaApiConfiguration.d.ts

@@ -0,0 +1,10 @@
+import { BaseUrlApiConfiguration } from "../../core/api/BaseUrlApiConfiguration.js";
+import { RetryFunction } from "../../core/api/RetryFunction.js";
+import { ThrottleFunction } from "../../core/api/ThrottleFunction.js";
+export declare class OllamaApiConfiguration extends BaseUrlApiConfiguration {
+    constructor({ baseUrl, retry, throttle, }?: {
+        baseUrl?: string;
+        retry?: RetryFunction;
+        throttle?: ThrottleFunction;
+    });
+}

package/model-provider/ollama/OllamaApiConfiguration.js

@@ -0,0 +1,11 @@
+import { BaseUrlApiConfiguration } from "../../core/api/BaseUrlApiConfiguration.js";
+export class OllamaApiConfiguration extends BaseUrlApiConfiguration {
+    constructor({ baseUrl = "http://127.0.0.1:11434", retry, throttle, } = {}) {
+        super({
+            baseUrl,
+            headers: {},
+            retry,
+            throttle,
+        });
+    }
+}