Add Together embedder to TS SDK (#5989)

Co-authored-by: kartik-mem0 <kartik.labhshetwar@mem0.ai>

docs/components/embedders/models/together.mdx

@@ -1,15 +1,20 @@
 ---
 title: Together
-description: "Configure Together AI as an embedding provider in Mem0 with support for 768-dimensional embedding models."
+description: "Configure Together AI as an embedding provider in Mem0 with support for 1024-dimensional embedding models."
 ---
 
-To use Together embedding models, set the `TOGETHER_API_KEY` environment variable. You can obtain the Together API key from the [Together Platform](https://api.together.xyz/settings/api-keys).
+To use Together embedding models, set the `TOGETHER_API_KEY` environment variable. You can obtain the Together API key from the [Together Platform](https://api.together.ai/settings/projects/~current/api-keys).
 
 ### Usage
 
-<Note> The `embedding_model_dims` parameter for `vector_store` should be set to `768` for Together embedder. </Note>
+<Note> The `embedding_model_dims` parameter for `vector_store` should be set to `1024` for Together embedder. </Note>
 
-```python
+<Warning>
+**Breaking default change.** The default Together embedding model is now `intfloat/multilingual-e5-large-instruct` (**1024-dim**), replacing the previous default `togethercomputer/m2-bert-80M-8k-retrieval` (**768-dim**). If you created a self-hosted vector store with the old default, its collection is 768-dim and will reject the new 1024-dim vectors **recreate/reindex the collection at 1024 dimensions** after upgrading. To defer the change, pin the previous values explicitly (`model="togethercomputer/m2-bert-80M-8k-retrieval"`, `embedding_dims=768`) note Together no longer lists this model among its recommended embeddings, so reindexing at 1024 is the durable path.
+</Warning>
+
+<CodeGroup>
+```python Python
 import os
 from mem0 import Memory
 
@@ -20,7 +25,7 @@ config = {
     "embedder": {
         "provider": "together",
         "config": {
-            "model": "togethercomputer/m2-bert-80M-8k-retrieval"
+            "model": "intfloat/multilingual-e5-large-instruct"
         }
     }
 }
@@ -29,18 +34,50 @@ m = Memory.from_config(config)
 messages = [
     {"role": "user", "content": "I'm planning to watch a movie tonight. Any recommendations?"},
     {"role": "assistant", "content": "How about thriller movies? They can be quite engaging."},
-    {"role": "user", "content": "I’m not a big fan of thriller movies but I love sci-fi movies."},
+    {"role": "user", "content": "I'm not a big fan of thriller movies but I love sci-fi movies."},
     {"role": "assistant", "content": "Got it! I'll avoid thriller recommendations and suggest sci-fi movies in the future."}
 ]
 m.add(messages, user_id="john")
 ```
 
+```typescript TypeScript
+import { Memory } from 'mem0ai/oss';
+
+const config = {
+  embedder: {
+    provider: 'together',
+    config: {
+      apiKey: process.env.TOGETHER_API_KEY || '',
+      model: 'intfloat/multilingual-e5-large-instruct',
+      embeddingDims: 1024,
+    },
+  },
+};
+
+const memory = new Memory(config);
+await memory.add("I'm visiting Paris", { userId: "john" });
+```
+
+</CodeGroup>
+
 ### Config
 
 Here are the parameters available for configuring Together embedder:
 
+<Tabs>
+<Tab title="Python">
 | Parameter | Description | Default Value |
 | --- | --- | --- |
-| `model` | The name of the embedding model to use | `togethercomputer/m2-bert-80M-8k-retrieval` |
+| `model` | The name of the embedding model to use | `intfloat/multilingual-e5-large-instruct` |
-| `embedding_dims` | Dimensions of the embedding model | `768` |
+| `embedding_dims` | Dimensions of the embedding model | `1024` |
 | `api_key` | The Together API key | `None` |
+</Tab>
+<Tab title="TypeScript">
+| Parameter | Description | Default Value |
+| --- | --- | --- |
+| `model` | The name of the embedding model to use | `intfloat/multilingual-e5-large-instruct` |
+| `embeddingDims` | Dimensions of the embedding model for vector store configuration | `1024` |
+| `apiKey` | The Together API key | `TOGETHER_API_KEY` |
+| `baseURL` | Base URL for an OpenAI-compatible Together endpoint | `https://api.together.ai/v1` |
+</Tab>
+</Tabs>

docs/components/embedders/overview.mdx

@@ -10,7 +10,7 @@ Mem0 offers support for various embedding models, allowing users to choose the o
 See the list of supported embedders below.
 
 <Note>
-  All embedders listed below are supported in the Python implementation. The TypeScript implementation supports: **OpenAI**, **Azure OpenAI**, **Google AI**, **Langchain**, **LM Studio**, and **Ollama**.
+  All embedders listed below are supported in the Python implementation. The TypeScript implementation supports: **OpenAI**, **Azure OpenAI**, **Google AI**, **Langchain**, **LM Studio**, **Ollama**, and **Together**.
 </Note>
 
 <CardGroup cols={4}>

docs/components/llms/models/together.mdx

@@ -3,7 +3,7 @@ title: Together
 description: "Configure Together AI as an LLM provider in Mem0 with API key setup and Mixtral model configuration."
 ---
 
-To use Together LLM models, you have to set the `TOGETHER_API_KEY` environment variable. You can obtain the Together API key from their [Account settings page](https://api.together.xyz/settings/api-keys).
+To use Together LLM models, you have to set the `TOGETHER_API_KEY` environment variable. You can obtain the Together API key from their [Account settings page](https://api.together.ai/settings/projects/~current/api-keys).
 
 ## Usage
 
@@ -37,4 +37,4 @@ m.add(messages, user_id="alice", metadata={"category": "movies"})
 
 ## Config
 
-All available parameters for the `together` config are present in [Master List of All Params in Config](../config).
+All available parameters for the `together` config are present in [Master List of All Params in Config](../config).

mem0-ts/src/oss/src/embeddings/together.ts

@@ -0,0 +1,24 @@
+import { OpenAIEmbedder } from "./openai";
+import { EmbeddingConfig } from "../types";
+
+const DEFAULT_BASE_URL = "https://api.together.ai/v1";
+const DEFAULT_MODEL = "intfloat/multilingual-e5-large-instruct";
+
+export class TogetherEmbedder extends OpenAIEmbedder {
+  constructor(config: EmbeddingConfig) {
+    const openAICompatibleConfig = { ...config };
+    delete openAICompatibleConfig.embeddingDims;
+
+    const apiKey = config.apiKey || process.env.TOGETHER_API_KEY;
+    if (!apiKey) {
+      throw new Error("Together API key is required");
+    }
+
+    super({
+      ...openAICompatibleConfig,
+      apiKey,
+      baseURL: config.baseURL || config.url || DEFAULT_BASE_URL,
+      model: config.model || DEFAULT_MODEL,
+    });
+  }
+}

mem0-ts/src/oss/src/index.ts

@@ -5,6 +5,7 @@ export * from "./embeddings/base";
 export * from "./embeddings/openai";
 export * from "./embeddings/ollama";
 export * from "./embeddings/lmstudio";
+export * from "./embeddings/together";
 export * from "./embeddings/google";
 export * from "./embeddings/azure";
 export * from "./embeddings/langchain";

mem0-ts/src/oss/src/utils/factory.ts

@@ -1,6 +1,7 @@
 import { OpenAIEmbedder } from "../embeddings/openai";
 import { OllamaEmbedder } from "../embeddings/ollama";
 import { LMStudioEmbedder } from "../embeddings/lmstudio";
+import { TogetherEmbedder } from "../embeddings/together";
 import { OpenAILLM } from "../llms/openai";
 import { OpenAIStructuredLLM } from "../llms/openai_structured";
 import { AnthropicLLM } from "../llms/anthropic";
@@ -51,6 +52,8 @@ export class EmbedderFactory {
         return new OllamaEmbedder(config);
       case "lmstudio":
         return new LMStudioEmbedder(config);
+      case "together":
+        return new TogetherEmbedder(config);
       case "google":
       case "gemini":
         return new GoogleEmbedder(config);

mem0-ts/src/oss/tests/factory.unit.test.ts

@@ -35,6 +35,11 @@ jest.mock("../src/embeddings/lmstudio", () => ({
     .fn()
     .mockImplementation((config) => ({ type: "lmstudio-embedder", config })),
 }));
+jest.mock("../src/embeddings/together", () => ({
+  TogetherEmbedder: jest
+    .fn()
+    .mockImplementation((config) => ({ type: "together-embedder", config })),
+}));
 
 jest.mock("../src/llms/openai", () => ({
   OpenAILLM: jest
@@ -190,6 +195,7 @@ describe("EmbedderFactory", () => {
     ["azure_openai"],
     ["langchain"],
     ["lmstudio"],
+    ["together"],
   ])("creates embedder for provider '%s'", (provider) => {
     expect(() =>
       EmbedderFactory.create(provider, dummyEmbedConfig),

mem0-ts/src/oss/tests/together-embedder.test.ts

@@ -0,0 +1,119 @@
+/// <reference types="jest" />
+
+const mockEmbeddingsCreate = jest.fn();
+const mockOpenAI = jest.fn().mockImplementation(() => ({
+  embeddings: { create: mockEmbeddingsCreate },
+}));
+
+jest.mock("openai", () => ({
+  __esModule: true,
+  default: mockOpenAI,
+}));
+
+import { TogetherEmbedder } from "../src/embeddings/together";
+
+const mockEmbedding = [0.1, 0.2, 0.3];
+const originalEnv = process.env;
+
+describe("TogetherEmbedder (unit)", () => {
+  beforeEach(() => {
+    jest.resetModules();
+    process.env = { ...originalEnv };
+    delete process.env.TOGETHER_API_KEY;
+    mockOpenAI.mockClear();
+    mockEmbeddingsCreate.mockReset();
+    mockEmbeddingsCreate.mockResolvedValue({
+      data: [{ index: 0, embedding: mockEmbedding }],
+    });
+  });
+
+  afterAll(() => {
+    process.env = originalEnv;
+  });
+
+  it("uses Together defaults with an API key from config", async () => {
+    const embedder = new TogetherEmbedder({ apiKey: "test-key" });
+
+    await embedder.embed("hello");
+
+    expect(mockOpenAI).toHaveBeenCalledWith({
+      apiKey: "test-key",
+      baseURL: "https://api.together.ai/v1",
+    });
+    expect(mockEmbeddingsCreate).toHaveBeenCalledWith({
+      model: "intfloat/multilingual-e5-large-instruct",
+      input: "hello",
+      encoding_format: "float",
+    });
+  });
+
+  it("uses TOGETHER_API_KEY when config apiKey is not provided", async () => {
+    process.env.TOGETHER_API_KEY = "env-key";
+
+    const embedder = new TogetherEmbedder({});
+
+    await embedder.embed("hello");
+
+    expect(mockOpenAI).toHaveBeenCalledWith({
+      apiKey: "env-key",
+      baseURL: "https://api.together.ai/v1",
+    });
+  });
+
+  it("supports custom model and baseURL without forwarding embeddingDims", async () => {
+    const embedder = new TogetherEmbedder({
+      apiKey: "test-key",
+      model: "custom-together-embed",
+      baseURL: "https://proxy.example.com/v1",
+      embeddingDims: 512,
+    });
+
+    await embedder.embed("hello");
+
+    expect(mockOpenAI).toHaveBeenCalledWith({
+      apiKey: "test-key",
+      baseURL: "https://proxy.example.com/v1",
+    });
+    expect(mockEmbeddingsCreate).toHaveBeenCalledWith({
+      model: "custom-together-embed",
+      input: "hello",
+      encoding_format: "float",
+    });
+  });
+
+  it("uses url as a baseURL fallback", async () => {
+    const embedder = new TogetherEmbedder({
+      apiKey: "test-key",
+      url: "https://url-fallback.example.com/v1",
+    });
+
+    await embedder.embed("hello");
+
+    expect(mockOpenAI).toHaveBeenCalledWith({
+      apiKey: "test-key",
+      baseURL: "https://url-fallback.example.com/v1",
+    });
+  });
+
+  it("sorts batch embeddings by response index", async () => {
+    mockEmbeddingsCreate.mockResolvedValueOnce({
+      data: [
+        { index: 1, embedding: [0.3, 0.4] },
+        { index: 0, embedding: [0.1, 0.2] },
+      ],
+    });
+
+    const embedder = new TogetherEmbedder({ apiKey: "test-key" });
+
+    await expect(embedder.embedBatch(["first", "second"])).resolves.toEqual([
+      [0.1, 0.2],
+      [0.3, 0.4],
+    ]);
+  });
+
+  it("throws when no API key is available", () => {
+    expect(() => new TogetherEmbedder({})).toThrow(
+      "Together API key is required",
+    );
+  });
+});

mem0/embeddings/together.py

@@ -11,10 +11,9 @@ class TogetherEmbedding(EmbeddingBase):
     def __init__(self, config: Optional[BaseEmbedderConfig] = None):
         super().__init__(config)
 
-        self.config.model = self.config.model or "togethercomputer/m2-bert-80M-8k-retrieval"
+        self.config.model = self.config.model or "intfloat/multilingual-e5-large-instruct"
         api_key = self.config.api_key or os.getenv("TOGETHER_API_KEY")
-        # TODO: check if this is correct
+        self.config.embedding_dims = self.config.embedding_dims or 1024
-        self.config.embedding_dims = self.config.embedding_dims or 768
         self.client = Together(api_key=api_key)
 
     def embed(self, text, memory_action: Optional[Literal["add", "search", "update"]] = None):

tests/embeddings/test_together_embeddings.py

@@ -5,6 +5,9 @@ import pytest
 from mem0.configs.embeddings.base import BaseEmbedderConfig
 from mem0.embeddings.together import TogetherEmbedding
 
+DEFAULT_MODEL = "intfloat/multilingual-e5-large-instruct"
+DEFAULT_EMBEDDING_DIMS = 1024
+
 
 @pytest.fixture
 def mock_together_client():
@@ -15,7 +18,7 @@ def mock_together_client():
 
 
 def test_embed_text(mock_together_client):
-    config = BaseEmbedderConfig(model="togethercomputer/m2-bert-80M-8k-retrieval", embedding_dims=768)
+    config = BaseEmbedderConfig(model=DEFAULT_MODEL, embedding_dims=DEFAULT_EMBEDDING_DIMS)
     embedder = TogetherEmbedding(config)
 
     mock_together_client.embeddings.create.return_value = Mock(data=[Mock(embedding=[0.1, 0.2, 0.3, 0.4, 0.5])])
@@ -23,14 +26,12 @@ def test_embed_text(mock_together_client):
     text = "Sample text to embed."
     embedding = embedder.embed(text)
 
-    mock_together_client.embeddings.create.assert_called_once_with(
+    mock_together_client.embeddings.create.assert_called_once_with(model=DEFAULT_MODEL, input=text)
-        model="togethercomputer/m2-bert-80M-8k-retrieval", input=text
-    )
     assert embedding == [0.1, 0.2, 0.3, 0.4, 0.5]
 
 
 def test_embed_batch_single_call(mock_together_client):
-    config = BaseEmbedderConfig(model="togethercomputer/m2-bert-80M-8k-retrieval", embedding_dims=768)
+    config = BaseEmbedderConfig(model=DEFAULT_MODEL, embedding_dims=DEFAULT_EMBEDDING_DIMS)
     embedder = TogetherEmbedding(config)
 
     mock_item0 = Mock(index=0, embedding=[0.1, 0.2, 0.3])
@@ -40,14 +41,12 @@ def test_embed_batch_single_call(mock_together_client):
     texts = ["First text.", "Second text."]
     embeddings = embedder.embed_batch(texts)
 
-    mock_together_client.embeddings.create.assert_called_once_with(
+    mock_together_client.embeddings.create.assert_called_once_with(model=DEFAULT_MODEL, input=texts)
-        model="togethercomputer/m2-bert-80M-8k-retrieval", input=texts
-    )
     assert embeddings == [[0.1, 0.2, 0.3], [0.4, 0.5, 0.6]]
 
 
 def test_embed_batch_empty_list(mock_together_client):
-    config = BaseEmbedderConfig(model="togethercomputer/m2-bert-80M-8k-retrieval", embedding_dims=768)
+    config = BaseEmbedderConfig(model=DEFAULT_MODEL, embedding_dims=DEFAULT_EMBEDDING_DIMS)
     embedder = TogetherEmbedding(config)
 
     result = embedder.embed_batch([])
@@ -57,7 +56,7 @@ def test_embed_batch_empty_list(mock_together_client):
 
 
 def test_embed_batch_count_mismatch_raises(mock_together_client):
-    config = BaseEmbedderConfig(model="togethercomputer/m2-bert-80M-8k-retrieval", embedding_dims=768)
+    config = BaseEmbedderConfig(model=DEFAULT_MODEL, embedding_dims=DEFAULT_EMBEDDING_DIMS)
     embedder = TogetherEmbedding(config)
 
     mock_item0 = Mock(index=0, embedding=[0.1, 0.2, 0.3])
@@ -65,3 +64,24 @@ def test_embed_batch_count_mismatch_raises(mock_together_client):
 
     with pytest.raises(ValueError, match="returned 1 embeddings for 2 texts"):
         embedder.embed_batch(["first text", "second text"])
+
+
+def test_default_config_applies_together_defaults(mock_together_client):
+    embedder = TogetherEmbedding(BaseEmbedderConfig())
+
+    assert embedder.config.model == DEFAULT_MODEL
+    assert embedder.config.embedding_dims == DEFAULT_EMBEDDING_DIMS
+
+
+def test_explicit_config_overrides_defaults(mock_together_client):
+    # The `config.x or default` wiring must honor user-provided values, not clobber them.
+    config = BaseEmbedderConfig(model="BAAI/bge-base-en-v1.5", embedding_dims=768)
+    embedder = TogetherEmbedding(config)
+
+    assert embedder.config.model == "BAAI/bge-base-en-v1.5"
+    assert embedder.config.embedding_dims == 768
+
+    # ...and the chosen model actually reaches the Together API call.
+    mock_together_client.embeddings.create.return_value = Mock(data=[Mock(embedding=[0.0] * 768)])
+    embedder.embed("hello")
+    mock_together_client.embeddings.create.assert_called_once_with(model="BAAI/bge-base-en-v1.5", input="hello")