Self-evolving Context Database for AI Agents. Unify Agent Memory, Knowledge RAG and Skills.
Python agent-contextagent-memoryagent-pluginsagentic-ragcontext-databasedsh-pluginknowledge-basememorymemory-managementself-evolvingvector-database
feat: add Jina AI embedding provider (#245)
Han Xiaoea2a508a9cd9
diff --git a/README.md b/README.mdindex ccd7f1d6f..bed8fb71a 100644--- a/README.md+++ b/README.md@@ -227,7 +227,7 @@ Create a configuration file `~/.openviking/ov.conf`:}```-> **Note**: For embedding models, currently only `volcengine` (Doubao) and `openai` providers are supported. For VLM models, we support multiple providers including volcengine, openai, deepseek, anthropic, gemini, moonshot, zhipu, dashscope, minimax, and more.+> **Note**: For embedding models, currently `volcengine` (Doubao), `openai`, and `jina` providers are supported. For VLM models, we support multiple providers including volcengine, openai, deepseek, anthropic, gemini, moonshot, zhipu, dashscope, minimax, and more.#### Configuration Examplesdiff --git a/README_CN.md b/README_CN.mdindex c09354ed6..b2f8875ff 100644--- a/README_CN.md+++ b/README_CN.md@@ -91,7 +91,7 @@ OpenViking 支持多种模型服务:"dense": {"api_base" : "<api-endpoint>", // API 服务端点地址"api_key" : "<your-api-key>", // 模型服务的 API 密钥- "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)+ "provider" : "<provider-type>", // 提供商类型(volcengine、openai 或 jina)"dimension": 1024, // 向量维度"model" : "<model-name>" // Embedding 模型名称(如 doubao-embedding-vision-250615 或 text-embedding-3-large)}@@ -99,7 +99,7 @@ OpenViking 支持多种模型服务:"vlm": {"api_base" : "<api-endpoint>", // API 服务端点地址"api_key" : "<your-api-key>", // 模型服务的 API 密钥- "provider" : "<provider-type>", // 提供商类型(volcengine 或 openai)+ "provider" : "<provider-type>", // 提供商类型(volcengine、openai 或 jina)"model" : "<model-name>" // VLM 模型名称(如 doubao-seed-1-8-251228 或 gpt-4-vision-preview)}}diff --git a/docs/en/api/01-overview.md b/docs/en/api/01-overview.mdindex 7b1b9c330..97cd52c35 100644--- a/docs/en/api/01-overview.md+++ b/docs/en/api/01-overview.md@@ -35,7 +35,7 @@ Minimal configuration example:"dense": {"api_base": "<api-endpoint>","api_key": "<your-api-key>",- "provider": "<volcengine|openai>",+ "provider": "<volcengine|openai|jina>","dimension": 1024,"model": "<model-name>"}@@ -43,7 +43,7 @@ Minimal configuration example:"vlm": {"api_base": "<api-endpoint>","api_key": "<your-api-key>",- "provider": "<volcengine|openai>",+ "provider": "<volcengine|openai|jina>","model": "<model-name>"}}diff --git a/docs/en/faq/faq.md b/docs/en/faq/faq.mdindex 4cc8a9f9d..2f15f3f2e 100644--- a/docs/en/faq/faq.md+++ b/docs/en/faq/faq.md@@ -109,6 +109,7 @@ Config files at the default path `~/.openviking/ov.conf` are loaded automaticall| `volcengine` | Volcengine Embedding API (Recommended) || `openai` | OpenAI Embedding API || `vikingdb` | VikingDB Embedding API |+| `jina` | Jina AI Embedding API |Supports Dense, Sparse, and Hybrid embedding modes.diff --git a/docs/en/guides/01-configuration.md b/docs/en/guides/01-configuration.mdindex 6442210e5..7c3c0fb2a 100644--- a/docs/en/guides/01-configuration.md+++ b/docs/en/guides/01-configuration.md@@ -118,7 +118,7 @@ Embedding model configuration for vector search, supporting dense, sparse, and h| Parameter | Type | Description ||-----------|------|-------------|-| `provider` | str | `"volcengine"`, `"openai"`, or `"vikingdb"` |+| `provider` | str | `"volcengine"`, `"openai"`, `"vikingdb"`, or `"jina"` || `api_key` | str | API key || `model` | str | Model name || `dimension` | int | Vector dimension |@@ -138,6 +138,7 @@ With `input: "multimodal"`, OpenViking can embed text, images (PNG, JPG, etc.),- `openai`: OpenAI Embedding API- `volcengine`: Volcengine Embedding API- `vikingdb`: VikingDB Embedding API+- `jina`: Jina AI Embedding API**vikingdb provider example:**@@ -156,6 +157,43 @@ With `input: "multimodal"`, OpenViking can embed text, images (PNG, JPG, etc.),}```+**jina provider example:**++```json+{+ "embedding": {+ "dense": {+ "provider": "jina",+ "api_key": "jina_xxx",+ "model": "jina-embeddings-v5-text-small",+ "dimension": 1024+ }+ }+}+```++Available Jina models:+- `jina-embeddings-v5-text-small`: 677M params, 1024 dim, max seq 32768 (default)+- `jina-embeddings-v5-text-nano`: 239M params, 768 dim, max seq 8192++Get your API key at https://jina.ai++**Local deployment (GGUF/MLX):** Jina embedding models are open-weight and available in GGUF and MLX formats on [Hugging Face](https://huggingface.co/jinaai). You can run them locally with any OpenAI-compatible server (e.g. llama.cpp, MLX, vLLM) and point the `api_base` to your local endpoint:++```json+{+ "embedding": {+ "dense": {+ "provider": "jina",+ "api_key": "local",+ "api_base": "http://localhost:8080/v1",+ "model": "jina-embeddings-v5-text-nano",+ "dimension": 768+ }+ }+}+```+#### Sparse Embedding```jsondiff --git a/docs/zh/api/01-overview.md b/docs/zh/api/01-overview.mdindex 1cbdbcf95..28dbf6a7e 100644--- a/docs/zh/api/01-overview.md+++ b/docs/zh/api/01-overview.md@@ -35,7 +35,7 @@ export OPENVIKING_CONFIG_FILE=/path/to/ov.conf"dense": {"api_base": "<api-endpoint>","api_key": "<your-api-key>",- "provider": "<volcengine|openai>",+ "provider": "<volcengine|openai|jina>","dimension": 1024,"model": "<model-name>"}@@ -43,7 +43,7 @@ export OPENVIKING_CONFIG_FILE=/path/to/ov.conf"vlm": {"api_base": "<api-endpoint>","api_key": "<your-api-key>",- "provider": "<volcengine|openai>",+ "provider": "<volcengine|openai|jina>","model": "<model-name>"}}diff --git a/docs/zh/faq/faq.md b/docs/zh/faq/faq.mdindex d3f7ba3b7..05d240e25 100644--- a/docs/zh/faq/faq.md+++ b/docs/zh/faq/faq.md@@ -109,6 +109,7 @@ pip install openviking| `volcengine` | 火山引擎 Embedding API(推荐) || `openai` | OpenAI Embedding API || `vikingdb` | VikingDB Embedding API |+| `jina` | Jina AI Embedding API |支持 Dense、Sparse 和 Hybrid 三种 Embedding 模式。diff --git a/docs/zh/guides/01-configuration.md b/docs/zh/guides/01-configuration.mdindex 7ee667626..ec70d4bd4 100644--- a/docs/zh/guides/01-configuration.md+++ b/docs/zh/guides/01-configuration.md@@ -120,7 +120,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支| 参数 | 类型 | 说明 ||------|------|------|-| `provider` | str | `"volcengine"`、`"openai"` 或 `"vikingdb"` |+| `provider` | str | `"volcengine"`、`"openai"`、`"vikingdb"` 或 `"jina"` || `api_key` | str | API Key || `model` | str | 模型名称 || `dimension` | int | 向量维度 |@@ -140,6 +140,7 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支- `openai`: OpenAI Embedding API- `volcengine`: 火山引擎 Embedding API- `vikingdb`: VikingDB Embedding API+- `jina`: Jina AI Embedding API**vikingdb provider 配置示例:**@@ -158,6 +159,43 @@ OpenViking 使用 JSON 配置文件(`ov.conf`)进行设置。配置文件支}```+**jina provider 配置示例:**++```json+{+ "embedding": {+ "dense": {+ "provider": "jina",+ "api_key": "jina_xxx",+ "model": "jina-embeddings-v5-text-small",+ "dimension": 1024+ }+ }+}+```++可用 Jina 模型:+- `jina-embeddings-v5-text-small`: 677M 参数, 1024 维, 最大序列长度 32768 (默认)+- `jina-embeddings-v5-text-nano`: 239M 参数, 768 维, 最大序列长度 8192++**本地部署 (GGUF/MLX):** Jina 嵌入模型是开源的, 在 [Hugging Face](https://huggingface.co/jinaai) 上提供 GGUF 和 MLX 格式。可以使用任何 OpenAI 兼容的推理服务器 (如 llama.cpp、MLX、vLLM) 本地运行, 并将 `api_base` 指向本地端点:++```json+{+ "embedding": {+ "dense": {+ "provider": "jina",+ "api_key": "local",+ "api_base": "http://localhost:8080/v1",+ "model": "jina-embeddings-v5-text-nano",+ "dimension": 768+ }+ }+}+```++获取 API Key: https://jina.ai+#### Sparse Embedding```jsondiff --git a/openviking/models/embedder/__init__.py b/openviking/models/embedder/__init__.pyindex 30f459590..9ce209caa 100644--- a/openviking/models/embedder/__init__.py+++ b/openviking/models/embedder/__init__.py@@ -11,6 +11,7 @@ Provides three embedder abstractions:Supported providers:- OpenAI: Dense only- Volcengine: Dense, Sparse, Hybrid+- Jina AI: Dense only"""from openviking.models.embedder.base import (@@ -21,6 +22,7 @@ from openviking.models.embedder.base import (HybridEmbedderBase,SparseEmbedderBase,)+from openviking.models.embedder.jina_embedders import JinaDenseEmbedderfrom openviking.models.embedder.openai_embedders import OpenAIDenseEmbedderfrom openviking.models.embedder.vikingdb_embedders import (VikingDBDenseEmbedder,@@ -41,6 +43,8 @@ __all__ = ["SparseEmbedderBase","HybridEmbedderBase","CompositeHybridEmbedder",+ # Jina AI implementations+ "JinaDenseEmbedder",# OpenAI implementations"OpenAIDenseEmbedder",# Volcengine implementationsdiff --git a/openviking/models/embedder/jina_embedders.py b/openviking/models/embedder/jina_embedders.pynew file mode 100644index 000000000..d0c9b3964--- /dev/null+++ b/openviking/models/embedder/jina_embedders.py@@ -0,0 +1,169 @@+# Copyright (c) 2026 Beijing Volcano Engine Technology Co., Ltd.+# SPDX-License-Identifier: Apache-2.0+"""Jina AI Embedder Implementation"""++from typing import Any, Dict, List, Optional++import openai++from openviking.models.embedder.base import (+ DenseEmbedderBase,+ EmbedResult,+)++# Default dimensions for Jina embedding models+JINA_MODEL_DIMENSIONS = {+ "jina-embeddings-v5-text-small": 1024, # 677M params, max seq 32768+ "jina-embeddings-v5-text-nano": 768, # 239M params, max seq 8192+}+++class JinaDenseEmbedder(DenseEmbedderBase):+ """Jina AI Dense Embedder Implementation++ Uses Jina AI embedding API via OpenAI-compatible client.+ Supports task-specific embeddings and Matryoshka dimension reduction.++ Example:+ >>> embedder = JinaDenseEmbedder(+ ... model_name="jina-embeddings-v5-text-small",+ ... api_key="jina_xxx",+ ... dimension=512,+ ... task="retrieval.query"+ ... )+ >>> result = embedder.embed("Hello world")+ >>> print(len(result.dense_vector))+ 512+ """++ def __init__(+ self,+ model_name: str = "jina-embeddings-v5-text-small",+ api_key: Optional[str] = None,+ api_base: Optional[str] = None,+ dimension: Optional[int] = None,+ task: Optional[str] = None,+ late_chunking: Optional[bool] = None,+ config: Optional[Dict[str, Any]] = None,+ ):+ """Initialize Jina AI Dense Embedder++ Args:+ model_name: Jina model name, defaults to jina-embeddings-v5-text-small+ api_key: API key, required+ api_base: API base URL, defaults to https://api.jina.ai/v1+ dimension: Dimension for Matryoshka reduction, optional+ task: Task type for task-specific embeddings, optional.+ Valid values: retrieval.query, retrieval.passage,+ text-matching, classification, separation+ late_chunking: Enable late chunking via extra_body, optional+ config: Additional configuration dict++ Raises:+ ValueError: If api_key is not provided+ """+ super().__init__(model_name, config)++ self.api_key = api_key+ self.api_base = api_base or "https://api.jina.ai/v1"+ self.dimension = dimension+ self.task = task+ self.late_chunking = late_chunking++ if not self.api_key:+ raise ValueError("api_key is required")++ # Initialize OpenAI-compatible client with Jina base URL+ self.client = openai.OpenAI(+ api_key=self.api_key,+ base_url=self.api_base,+ )++ # Determine dimension+ max_dim = JINA_MODEL_DIMENSIONS.get(model_name, 1024)+ if dimension is not None and dimension > max_dim:+ raise ValueError(+ f"Requested dimension {dimension} exceeds maximum {max_dim} for model '{model_name}'. "+ f"Jina models support Matryoshka dimension reduction up to {max_dim}."+ )+ self._dimension = dimension if dimension is not None else max_dim++ def _build_extra_body(self) -> Optional[Dict[str, Any]]:+ """Build extra_body dict for Jina-specific parameters"""+ extra_body = {}+ if self.task is not None:+ extra_body["task"] = self.task+ if self.late_chunking is not None:+ extra_body["late_chunking"] = self.late_chunking+ return extra_body if extra_body else None++ def embed(self, text: str) -> EmbedResult:+ """Perform dense embedding on text++ Args:+ text: Input text++ Returns:+ EmbedResult: Result containing only dense_vector++ Raises:+ RuntimeError: When API call fails+ """+ try:+ kwargs: Dict[str, Any] = {"input": text, "model": self.model_name}+ if self.dimension:+ kwargs["dimensions"] = self.dimension++ extra_body = self._build_extra_body()+ if extra_body:+ kwargs["extra_body"] = extra_body++ response = self.client.embeddings.create(**kwargs)+ vector = response.data[0].embedding++ return EmbedResult(dense_vector=vector)+ except openai.APIError as e:+ raise RuntimeError(f"Jina API error: {e.message}") from e+ except Exception as e:+ raise RuntimeError(f"Embedding failed: {str(e)}") from e++ def embed_batch(self, texts: List[str]) -> List[EmbedResult]:+ """Batch embedding (Jina native support)++ Args:+ texts: List of texts++ Returns:+ List[EmbedResult]: List of embedding results++ Raises:+ RuntimeError: When API call fails+ """+ if not texts:+ return []++ try:+ kwargs: Dict[str, Any] = {"input": texts, "model": self.model_name}+ if self.dimension:+ kwargs["dimensions"] = self.dimension++ extra_body = self._build_extra_body()+ if extra_body:+ kwargs["extra_body"] = extra_body++ response = self.client.embeddings.create(**kwargs)++ return [EmbedResult(dense_vector=item.embedding) for item in response.data]+ except openai.APIError as e:+ raise RuntimeError(f"Jina API error: {e.message}") from e+ except Exception as e:+ raise RuntimeError(f"Batch embedding failed: {str(e)}") from e++ def get_dimension(self) -> int:+ """Get embedding dimension++ Returns:+ int: Vector dimension+ """+ return self._dimension+diff --git a/openviking_cli/utils/config/embedding_config.py b/openviking_cli/utils/config/embedding_config.pyindex e86e9ed31..ba9aab96e 100644--- a/openviking_cli/utils/config/embedding_config.py+++ b/openviking_cli/utils/config/embedding_config.py@@ -15,7 +15,7 @@ class EmbeddingModelConfig(BaseModel):batch_size: int = Field(default=32, description="Batch size for embedding generation")input: str = Field(default="multimodal", description="Input type: 'text' or 'multimodal'")provider: Optional[str] = Field(- default="volcengine", description="Provider type: 'openai', 'volcengine', 'vikingdb'"+ default="volcengine", description="Provider type: 'openai', 'volcengine', 'vikingdb', 'jina'")backend: Optional[str] = Field(default="volcengine",@@ -52,9 +52,9 @@ class EmbeddingModelConfig(BaseModel):if not self.provider:raise ValueError("Embedding provider is required")- if self.provider not in ["openai", "volcengine", "vikingdb"]:+ if self.provider not in ["openai", "volcengine", "vikingdb", "jina"]:raise ValueError(- f"Invalid embedding provider: '{self.provider}'. Must be one of: 'openai', 'volcengine', 'vikingdb'"+ f"Invalid embedding provider: '{self.provider}'. Must be one of: 'openai', 'volcengine', 'vikingdb', 'jina'")# Provider-specific validation@@ -80,6 +80,10 @@ class EmbeddingModelConfig(BaseModel):f"VikingDB provider requires the following fields: {', '.join(missing)}")+ elif self.provider == "jina":+ if not self.api_key:+ raise ValueError("Jina provider requires 'api_key' to be set")+return self@@ -125,6 +129,7 @@ class EmbeddingConfig(BaseModel):ValueError: If provider/type combination is not supported"""from openviking.models.embedder import (+ JinaDenseEmbedder,OpenAIDenseEmbedder,VikingDBDenseEmbedder,VikingDBHybridEmbedder,@@ -210,6 +215,15 @@ class EmbeddingConfig(BaseModel):"input_type": cfg.input,},),+ ("jina", "dense"): (+ JinaDenseEmbedder,+ lambda cfg: {+ "model_name": cfg.model,+ "api_key": cfg.api_key,+ "api_base": cfg.api_base,+ "dimension": cfg.dimension,+ },+ ),}key = (provider, embedder_type)diff --git a/tests/unit/__init__.py b/tests/unit/__init__.pynew file mode 100644index 000000000..819b3465d--- /dev/null+++ b/tests/unit/__init__.py@@ -0,0 +1,2 @@+# Copyright (c) 2026 Beijing Volcano Engine Technology Co., Ltd.+# SPDX-License-Identifier: Apache-2.0diff --git a/tests/unit/test_jina_embedder.py b/tests/unit/test_jina_embedder.pynew file mode 100644index 000000000..67a926681--- /dev/null+++ b/tests/unit/test_jina_embedder.py@@ -0,0 +1,271 @@+# Copyright (c) 2026 Beijing Volcano Engine Technology Co., Ltd.+# SPDX-License-Identifier: Apache-2.0+"""Tests for Jina AI Embedder"""++from unittest.mock import MagicMock, patch++import pytest++from openviking.models.embedder import JinaDenseEmbedder+from openviking.models.embedder.jina_embedders import (+ JINA_MODEL_DIMENSIONS,+)+++class TestJinaDenseEmbedder:+ """Test cases for JinaDenseEmbedder"""++ def test_init_requires_api_key(self):+ """Test that api_key is required"""+ with pytest.raises(ValueError, match="api_key is required"):+ JinaDenseEmbedder(model_name="jina-embeddings-v5-text-small")++ def test_init_with_api_key(self):+ """Test initialization with api_key"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ assert embedder.api_key == "test-api-key"+ assert embedder.model_name == "jina-embeddings-v5-text-small"+ assert embedder.api_base == "https://api.jina.ai/v1"++ def test_init_with_custom_api_base(self):+ """Test initialization with custom api_base"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ api_base="https://custom.api.jina.ai/v1",+ )+ assert embedder.api_base == "https://custom.api.jina.ai/v1"++ def test_default_dimension_v5_small(self):+ """Test default dimension for jina-embeddings-v5-text-small"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ assert embedder.get_dimension() == 1024++ def test_default_dimension_v5_nano(self):+ """Test default dimension for jina-embeddings-v5-text-nano"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-nano",+ api_key="test-api-key",+ )+ assert embedder.get_dimension() == 768++ def test_custom_dimension(self):+ """Test custom dimension for Matryoshka reduction"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ dimension=256,+ )+ assert embedder.get_dimension() == 256++ def test_model_dimensions_constant(self):+ """Test JINA_MODEL_DIMENSIONS constant"""+ assert "jina-embeddings-v5-text-small" in JINA_MODEL_DIMENSIONS+ assert "jina-embeddings-v5-text-nano" in JINA_MODEL_DIMENSIONS+ assert JINA_MODEL_DIMENSIONS["jina-embeddings-v5-text-small"] == 1024+ assert JINA_MODEL_DIMENSIONS["jina-embeddings-v5-text-nano"] == 768++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_single_text(self, mock_openai_class):+ """Test embedding a single text"""+ # Setup mock+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_embedding = MagicMock()+ mock_embedding.embedding = [0.1] * 1024++ mock_response = MagicMock()+ mock_response.data = [mock_embedding]+ mock_client.embeddings.create.return_value = mock_response++ # Create embedder and embed+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ result = embedder.embed("Hello world")++ # Verify+ assert result.dense_vector is not None+ assert len(result.dense_vector) == 1024+ mock_client.embeddings.create.assert_called_once()++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_with_dimension(self, mock_openai_class):+ """Test embedding with custom dimension parameter"""+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_embedding = MagicMock()+ mock_embedding.embedding = [0.1] * 768++ mock_response = MagicMock()+ mock_response.data = [mock_embedding]+ mock_client.embeddings.create.return_value = mock_response++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ dimension=768,+ )+ embedder.embed("Hello world")++ # Check dimensions was passed+ call_kwargs = mock_client.embeddings.create.call_args[1]+ assert call_kwargs["dimensions"] == 768++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_with_task(self, mock_openai_class):+ """Test embedding with task parameter"""+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_embedding = MagicMock()+ mock_embedding.embedding = [0.1] * 1024++ mock_response = MagicMock()+ mock_response.data = [mock_embedding]+ mock_client.embeddings.create.return_value = mock_response++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ task="retrieval.query",+ )+ embedder.embed("Hello world")++ # Check extra_body was passed with task+ call_kwargs = mock_client.embeddings.create.call_args[1]+ assert "extra_body" in call_kwargs+ assert call_kwargs["extra_body"]["task"] == "retrieval.query"++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_with_late_chunking(self, mock_openai_class):+ """Test embedding with late_chunking parameter"""+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_embedding = MagicMock()+ mock_embedding.embedding = [0.1] * 1024++ mock_response = MagicMock()+ mock_response.data = [mock_embedding]+ mock_client.embeddings.create.return_value = mock_response++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ late_chunking=True,+ )+ embedder.embed("Hello world")++ # Check extra_body was passed with late_chunking+ call_kwargs = mock_client.embeddings.create.call_args[1]+ assert "extra_body" in call_kwargs+ assert call_kwargs["extra_body"]["late_chunking"] is True++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_batch(self, mock_openai_class):+ """Test batch embedding"""+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_embeddings = [MagicMock(embedding=[0.1] * 1024) for _ in range(3)]++ mock_response = MagicMock()+ mock_response.data = mock_embeddings+ mock_client.embeddings.create.return_value = mock_response++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ results = embedder.embed_batch(["Hello", "World", "Test"])++ assert len(results) == 3+ for result in results:+ assert result.dense_vector is not None+ assert len(result.dense_vector) == 1024++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_batch_empty(self, mock_openai_class):+ """Test batch embedding with empty list"""+ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ results = embedder.embed_batch([])++ assert results == []+ mock_client.embeddings.create.assert_not_called()++ @patch("openviking.models.embedder.jina_embedders.openai.OpenAI")+ def test_embed_api_error(self, mock_openai_class):+ """Test embedding with API error"""+ import openai++ mock_client = MagicMock()+ mock_openai_class.return_value = mock_client++ mock_client.embeddings.create.side_effect = openai.APIError(+ message="Test API error",+ request=MagicMock(),+ body=None,+ )++ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )++ with pytest.raises(RuntimeError, match="Jina API error"):+ embedder.embed("Hello world")++ def test_build_extra_body_none(self):+ """Test _build_extra_body returns None when no params set"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ )+ assert embedder._build_extra_body() is None++ def test_build_extra_body_with_params(self):+ """Test _build_extra_body returns dict with params"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-small",+ api_key="test-api-key",+ task="retrieval.passage",+ late_chunking=True,+ )+ extra_body = embedder._build_extra_body()+ assert extra_body is not None+ assert extra_body["task"] == "retrieval.passage"+ assert extra_body["late_chunking"] is True++ def test_dimension_validation_exceeds_max(self):+ """Test that requesting dimension exceeding model max raises ValueError"""+ with pytest.raises(ValueError, match="exceeds maximum"):+ JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-nano",+ api_key="test-key",+ dimension=1024, # nano max is 768+ )++ def test_dimension_validation_within_range(self):+ """Test that requesting dimension within model max works"""+ embedder = JinaDenseEmbedder(+ model_name="jina-embeddings-v5-text-nano",+ api_key="test-key",+ dimension=256,+ )+ assert embedder.get_dimension() == 256