feat: add new llm and embedding provider JiekouAI

README.md

@@ -96,7 +96,7 @@ result = query("Write a report about xxx.") # Your question here
 #### LLM Configuration
 
 <pre><code>config.set_provider_config("llm", "(LLMName)", "(Arguments dict)")</code></pre>
-<p>The "LLMName" can be one of the following: ["DeepSeek", "OpenAI", "XAI", "SiliconFlow", "Aliyun", "PPIO", "TogetherAI", "Gemini", "Ollama", "Novita"]</p>
+<p>The "LLMName" can be one of the following: ["DeepSeek", "OpenAI", "XAI", "SiliconFlow", "Aliyun", "PPIO", "TogetherAI", "Gemini", "Ollama", "Novita", "Jiekou.AI"]</p>
 <p> The "Arguments dict" is a dictionary that contains the necessary arguments for the LLM class.</p>
 
 <details>
@@ -173,6 +173,13 @@ result = query("Write a report about xxx.") # Your question here
     <p> More details about PPIO: https://ppinfra.com/docs/get-started/quickstart.html?utm_source=github_deep-searcher </p>
 </details>
 
+<details>
+  <summary>Example (Claude Sonnet 4.5 from Jiekou.AI)</summary>
+    <p> Make sure you have prepared your Jiekou.AI API KEY as an env variable <code>JIEKOU_API_KEY</code>. You can create an API Key <a href="https://jiekou.ai/settings/key-management?utm_source=github_deep-searcher">here</a>. </p>
+    <pre><code>config.set_provider_config("llm", "JiekouAI", {"model": "claude-sonnet-4-5-20250929"})</code></pre>
+    <p> More details about Jiekou.AI: https://docs.jiekou.ai/docs/support/quickstart?utm_source=github_deep-searcher </p>
+</details>
+
 <details>
   <summary>Example (Ollama)</summary>
   <p> Follow <a href="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/jmorganca/ollama">these instructions</a> to set up and run a local Ollama instance:</p>
@@ -216,7 +223,7 @@ result = query("Write a report about xxx.") # Your question here
 
 #### Embedding Model Configuration
 <pre><code>config.set_provider_config("embedding", "(EmbeddingModelName)", "(Arguments dict)")</code></pre>
-<p>The "EmbeddingModelName" can be one of the following: ["MilvusEmbedding", "OpenAIEmbedding", "VoyageEmbedding", "SiliconflowEmbedding", "PPIOEmbedding", "NovitaEmbedding"]</p>
+<p>The "EmbeddingModelName" can be one of the following: ["MilvusEmbedding", "OpenAIEmbedding", "VoyageEmbedding", "SiliconflowEmbedding", "PPIOEmbedding", "NovitaEmbedding", "JiekouAIEmbedding"]</p>
 <p> The "Arguments dict" is a dictionary that contains the necessary arguments for the embedding model class.</p>
 
 <details>
@@ -308,6 +315,12 @@ result = query("Write a report about xxx.") # Your question here
     <p> More details about PPIO: https://ppinfra.com/docs/get-started/quickstart.html?utm_source=github_deep-searcher </p>
 </details>
 
+<details>
+  <summary>Example (Jiekou.AI embedding)</summary>
+    <p> Make sure you have prepared your Jiekou.AI API KEY as an env variable <code>JIEKOU_API_KEY</code>.</p>
+    <pre><code>config.set_provider_config("embedding", "JiekouAIEmbedding", {"model": "qwen/qwen3-embedding-8b"})</code></pre>
+    <p> More details about Jiekou.AI: https://docs.jiekou.ai/docs/support/quickstart?utm_source=github_deep-searcher </p>
+</details>
 
 <details>
   <summary>Example (FastEmbed embedding)</summary>
@@ -548,6 +561,7 @@ nest_asyncio.apply()
 - [PPIO](https://ppinfra.com/model-api/product/llm-api?utm_source=github_deep-searcher) (`PPIO_API_KEY` env variable required)
 - [Novita AI](https://novita.ai/docs/api-reference/model-apis-llm-create-embeddings?utm_source=github_deep-searcher&utm_medium=github_readme&utm_campaign=link) (`NOVITA_API_KEY` env variable required)
 - [IBM watsonx.ai](https://www.ibm.com/products/watsonx-ai/foundation-models#ibmembedding) (`WATSONX_APIKEY`, `WATSONX_URL`, `WATSONX_PROJECT_ID` env variables required)
+- [Jiekou.AI](https://jiekou.ai/?utm_source=github_deep-searcher) (`JIEKOU_API_KEY` env variable required)
 
 ### 🔹 LLM Support
 - [OpenAI](https://platform.openai.com/docs/models) (`OPENAI_API_KEY` env variable required)
@@ -562,6 +576,7 @@ nest_asyncio.apply()
 - [Ollama](https://ollama.com/)
 - [Novita AI](https://novita.ai/docs/guides/introduction?utm_source=github_deep-searcher&utm_medium=github_readme&utm_campaign=link) (`NOVITA_API_KEY` env variable required)
 - [IBM watsonx.ai](https://www.ibm.com/products/watsonx-ai/foundation-models#ibmfm) (`WATSONX_APIKEY`, `WATSONX_URL`, `WATSONX_PROJECT_ID` env variable required)
+- [Jiekou.AI](https://jiekou.ai/?utm_source=github_deep-searcher) (`JIEKOU_API_KEY` env variable required)
 
 ### 🔹 Document Loader
 - Local File

deepsearcher/config.yaml

@@ -24,6 +24,12 @@ provide_settings:
 ##      api_key: "sk_xxxxxx"  # Uncomment to override the `PPIO_API_KEY` set in the environment variable
 ##      base_url: ""
 
+#    provider: "Jiekou.AI"
+#    config:
+#      model: "claude-sonnet-4-5-20250929"
+##      api_key: "xxxx"  # Uncomment to override the `JIEKOU_API_KEY` set in the environment variable
+##      base_url: ""
+
 #    provider: "TogetherAI"
 #    config:
 #      model: "deepseek-ai/DeepSeek-R1"

deepsearcher/embedding/__init__.py

@@ -2,6 +2,7 @@
 from .fastembed_embdding import FastEmbedEmbedding
 from .gemini_embedding import GeminiEmbedding
 from .glm_embedding import GLMEmbedding
+from .jiekouai_embedding import JiekouAIEmbedding
 from .milvus_embedding import MilvusEmbedding
 from .novita_embedding import NovitaEmbedding
 from .ollama_embedding import OllamaEmbedding
@@ -28,4 +29,5 @@
     "NovitaEmbedding",
     "SentenceTransformerEmbedding",
     "WatsonXEmbedding",
+    "JiekouAIEmbedding",
 ]

deepsearcher/embedding/jiekouai_embedding.py

@@ -0,0 +1,131 @@
+import os
+from typing import List, Union
+
+import requests
+
+from deepsearcher.embedding.base import BaseEmbedding
+
+# TODO: Update with actual Jiekou.AI model dimensions when available
+JIEKOUAI_MODEL_DIM_MAP = {
+    "qwen/qwen3-embedding-0.6b": 1024,
+    "qwen/qwen3-embedding-8b": 1024,
+    "baai/bge-m3": 1024,
+}
+
+JIEKOUAI_EMBEDDING_API = "https://api.jiekou.ai/openai/v1/embeddings"
+
+
+class JiekouAIEmbedding(BaseEmbedding):
+    """
+    Jiekou.AI embedding model implementation.
+
+    This class provides an interface to the Jiekou.AI embedding API, which offers
+    various embedding models for text processing.
+    """
+
+    def __init__(self, model="qwen/qwen3-embedding-8b", batch_size=32, **kwargs):
+        """
+        Initialize the Jiekou.AI embedding model.
+
+        Args:
+            model (str): The model identifier to use for embeddings. Default is "baai/bge-m3".
+            batch_size (int): Maximum number of texts to process in a single batch. Default is 32.
+            **kwargs: Additional keyword arguments.
+                - api_key (str, optional): The Jiekou.AI API key. If not provided,
+                  it will be read from the JIEKOU_API_KEY environment variable.
+                - model_name (str, optional): Alternative way to specify the model.
+
+        Raises:
+            RuntimeError: If no API key is provided or found in environment variables.
+        """
+        if "model_name" in kwargs and (not model or model == "qwen/qwen3-embedding-8b"):
+            model = kwargs.pop("model_name")
+        self.model = model
+
+        if "api_key" in kwargs:
+            api_key = kwargs.pop("api_key")
+        else:
+            api_key = os.getenv("JIEKOU_API_KEY")
+
+        if not api_key or len(api_key) == 0:
+            raise RuntimeError("api_key is required for JiekouAIEmbedding")
+        self.api_key = api_key
+        self.batch_size = batch_size
+
+    def embed_query(self, text: str) -> List[float]:
+        """
+        Embed a single query text.
+
+        Args:
+            text (str): The query text to embed.
+
+        Returns:
+            List[float]: A list of floats representing the embedding vector.
+        """
+        return self._embed_input(text)[0]
+
+    def embed_documents(self, texts: List[str]) -> List[List[float]]:
+        """
+        Embed a list of document texts.
+
+        This method handles batching of document embeddings based on the configured
+        batch size to optimize API calls.
+
+        Args:
+            texts (List[str]): A list of document texts to embed.
+
+        Returns:
+            List[List[float]]: A list of embedding vectors, one for each input text.
+        """
+        # batch embedding
+        if self.batch_size > 0:
+            if len(texts) > self.batch_size:
+                batch_texts = [
+                    texts[i : i + self.batch_size] for i in range(0, len(texts), self.batch_size)
+                ]
+                embeddings = []
+                for batch_text in batch_texts:
+                    batch_embeddings = self._embed_input(batch_text)
+                    embeddings.extend(batch_embeddings)
+                return embeddings
+            return self._embed_input(texts)
+        return [self.embed_query(text) for text in texts]
+
+    def _embed_input(self, input: Union[str, List[str]]) -> List[List[float]]:
+        """
+        Internal method to handle the API call for embedding inputs.
+
+        Args:
+            input (Union[str, List[str]]): Either a single text string or a list of text strings to embed.
+
+        Returns:
+            List[List[float]]: A list of embedding vectors for the input(s).
+
+        Raises:
+            HTTPError: If the API request fails.
+        """
+        headers = {
+            "Authorization": f"Bearer {self.api_key}",
+            "Content-Type": "application/json",
+        }
+
+        # Handle both single string and list of strings
+        input_list = input if isinstance(input, list) else [input]
+
+        payload = {"model": self.model, "input": input_list}
+
+        response = requests.request("POST", JIEKOUAI_EMBEDDING_API, json=payload, headers=headers)
+        response.raise_for_status()
+        result = response.json()["data"]
+        sorted_results = sorted(result, key=lambda x: x["index"])
+        return [res["embedding"] for res in sorted_results]
+
+    @property
+    def dimension(self) -> int:
+        """
+        Get the dimensionality of the embeddings for the current model.
+
+        Returns:
+            int: The number of dimensions in the embedding vectors.
+        """
+        return JIEKOUAI_MODEL_DIM_MAP[self.model]

deepsearcher/llm/__init__.py

@@ -5,6 +5,7 @@
 from .deepseek import DeepSeek
 from .gemini import Gemini
 from .glm import GLM
+from .jiekouai import JiekouAI
 from .novita import Novita
 from .ollama import Ollama
 from .openai_llm import OpenAI
@@ -32,4 +33,5 @@
     "Novita",
     "Aliyun",
     "WatsonX",
+    "JiekouAI",
 ]

deepsearcher/llm/jiekouai.py

@@ -0,0 +1,61 @@
+import os
+from typing import Dict, List
+
+from deepsearcher.llm.base import BaseLLM, ChatResponse
+
+
+class JiekouAI(BaseLLM):
+    """
+    Jiekou.AI language model implementation.
+
+    This class provides an interface to interact with language models
+    hosted on the Jiekou.AI platform.
+
+    Attributes:
+        model (str): The model identifier to use on Jiekou.AI platform.
+        client: The OpenAI-compatible client instance for Jiekou.AI API.
+    """
+
+    def __init__(self, model: str = "claude-sonnet-4-5-20250929", **kwargs):
+        """
+        Initialize a Jiekou.AI language model client.
+
+        Args:
+            model (str, optional): The model identifier to use. Defaults to "claude-sonnet-4-5-20250929".
+            **kwargs: Additional keyword arguments to pass to the OpenAI client.
+                - api_key: Jiekou.AI API key. If not provided, uses JIEKOU_API_KEY environment variable.
+                - base_url: Jiekou.AI API base URL. If not provided, defaults to "https://api.jiekou.ai/openai/v1".
+        """
+        from openai import OpenAI as OpenAI_
+
+        self.model = model
+        if "api_key" in kwargs:
+            api_key = kwargs.pop("api_key")
+        else:
+            api_key = os.getenv("JIEKOU_API_KEY")
+        if "base_url" in kwargs:
+            base_url = kwargs.pop("base_url")
+        else:
+            base_url = "https://api.jiekou.ai/openai/v1"
+        self.client = OpenAI_(api_key=api_key, base_url=base_url, **kwargs)
+
+    def chat(self, messages: List[Dict]) -> ChatResponse:
+        """
+        Send a chat message to the Jiekou.AI model and get a response.
+
+        Args:
+            messages (List[Dict]): A list of message dictionaries, typically in the format
+                                  [{"role": "system", "content": "..."},
+                                   {"role": "user", "content": "..."}]
+
+        Returns:
+            ChatResponse: An object containing the model's response and token usage information.
+        """
+        completion = self.client.chat.completions.create(
+            model=self.model,
+            messages=messages,
+        )
+        return ChatResponse(
+            content=completion.choices[0].message.content,
+            total_tokens=completion.usage.total_tokens,
+        )

docs/configuration/embedding.md

@@ -25,6 +25,7 @@ config.set_provider_config("embedding", "(EmbeddingModelName)", "(Arguments dict
 | **NovitaEmbedding** | Novita AI embedding | Cost-effective |
 | **SentenceTransformerEmbedding** | Sentence Transfomer Embedding | Self-hosted option |
 | **IBM watsonx.ai** | Various options | IBM's Enterprise AI platform |
+| **JiekouAIEmbedding** | Jiekou.AI embedding | High quality, cost-effective |
 
 ## 🔍 Provider Examples
 
@@ -124,3 +125,10 @@ config.set_provider_config("embedding", "VoyageEmbedding", {"model": "voyage-3"}
     config.set_provider_config("embedding", "WatsonXEmbedding", {"model": "ibm/slate-125m-english-rtrvr-v2"})
     ```
     *Requires `pip install ibm-watsonx-ai`*
+
+??? example "Jiekou.AI"
+
+    ```python
+    config.set_provider_config("embedding", "JiekouAIEmbedding", {"model": "qwen/qwen3-embedding-8b"})
+    ```
+    *Requires `JIEKOU_API_KEY` environment variable*

docs/configuration/llm.md

@@ -26,6 +26,7 @@ config.set_provider_config("llm", "(LLMName)", "(Arguments dict)")
 | **Bedrock** | Amazon Bedrock LLMs | anthropic.claude, ai21.j2 |
 | **Novita** | Novita AI models | Various options |
 | **IBM watsonx.ai** | IBM Enterprise AI platform | Various options |
+| **Jiekou.AI** | Jiekou.AI models | Claude, OpenAI, Deepseek, Grok, Qwen, etc. |
 
 ## 🔍 Provider Examples
 
@@ -190,3 +191,13 @@ config.set_provider_config("llm", "WatsonX", {"model": "ibm/granite-3-3-8b-instr
 
     More details about WatsonX: [https://www.ibm.com/products/watsonx-ai/foundation-models](https://www.ibm.com/products/watsonx-ai/foundation-models)
 ```
+
+??? example "Jiekou.AI"
+
+    ```python
+    config.set_provider_config("llm", "JiekouAI", {"model": "claude-sonnet-4-5-20250929"})
+    ```
+    *Requires `JIEKOU_API_KEY` environment variable*
+
+    More details about Jiekou.AI: [https://docs.jiekou.ai/docs/support/quickstart](https://docs.jiekou.ai/docs/support/quickstart)
+

docs/integrations/index.md

@@ -25,6 +25,7 @@ Support for various embedding models to convert text into vector representations
 | **[PPIO](https://ppinfra.com/model-api/product/llm-api)** | `PPIO_API_KEY` | Flexible cloud embeddings |
 | **[Novita AI](https://novita.ai/docs/api-reference/model-apis-llm-create-embeddings)** | `NOVITA_API_KEY` | Rich model selection |
 | **[IBM watsonx.ai](https://www.ibm.com/products/watsonx-ai/foundation-models#ibmembedding)** | `WATSONX_APIKEY`, `WATSONX_URL`, `WATSONX_PROJECT_ID` | IBM's Enterprise AI platform |
+| **[Jiekou.AI](https://jiekou.ai/?utm_source=github_deep-searcher)** | `JIEKOU_API_KEY` | Jiekou.AI's embedding models |
 
 ## 🧠 Large Language Models {#llm-support}
 
@@ -44,6 +45,7 @@ Support for various large language models (LLMs) to process queries and generate
 | **[Ollama](https://ollama.com/)** | None | Local LLM deployment |
 | **[Novita AI](https://novita.ai/docs/guides/introduction)** | `NOVITA_API_KEY` | Diverse AI services |
 | **[IBM watsonx.ai](https://www.ibm.com/products/watsonx-ai/foundation-models#ibmfm)** | `WATSONX_APIKEY`, `WATSONX_URL`, `WATSONX_PROJECT_ID` | IBM's Enterprise AI platform |
+| **[Jiekou.AI](https://jiekou.ai/?utm_source=github_deep-searcher)** | `JIEKOU_API_KEY` | Various open-source and closed-source models |
 
 ## 📄 Document Loader {#document-loader}
 

env.example

@@ -12,6 +12,9 @@ SILICONFLOW_API_KEY=your_siliconflow_api_key_here
 # PPIO API Key (required if using PPIO models)
 PPIO_API_KEY=your_ppio_api_key_here
 
+# Jiekou.AI API Key (required if using Jiekou.AI models)
+JIEKOU_API_KEY=your_jiekouai_api_key_here
+
 # Together AI API Key (required if using Together AI models)
 TOGETHER_API_KEY=your_together_api_key_here
 

examples/basic_example.py

@@ -1,5 +1,6 @@
 import logging
 import os
+import glob
 
 from deepsearcher.offline_loading import load_from_local_files
 from deepsearcher.online_query import query
@@ -11,6 +12,8 @@
 current_dir = os.path.dirname(os.path.abspath(__file__))
 
 config = Configuration()  # Customize your config here
+config.set_provider_config("llm", "JiekouAI", {"model": "claude-sonnet-4-5-20250929"})
+config.set_provider_config("embedding", "JiekouAIEmbedding", {"model": "qwen/qwen3-embedding-8b"})
 init_config(config=config)
 
 
@@ -23,13 +26,15 @@
 
 # Hint: You can also load a single file, please execute it in the root directory of the deep searcher project
 load_from_local_files(
-    paths_or_directory=os.path.join(current_dir, "data/WhatisMilvus.pdf"),
-    collection_name="milvus_docs",
-    collection_description="All Milvus Documents",
-    # force_new_collection=True, # If you want to drop origin collection and create a new collection every time, set force_new_collection to True
+    paths_or_directory='/Users/jason/Documents/work/PPLabs/Platform/deep-searcher/examples/data/car_company_data.pdf',
+    collection_name="car_company_data",
+    collection_description="car_company_data",
+    force_new_collection=True, # If you want to drop origin collection and create a new collection every time, set force_new_collection to True
 )
 
-question = "Write a report comparing Milvus with other vector databases."
+question = """请从财务和宏观经济角度，对 A 股新能源汽车行业以及行业内 TOP5 车企的发展进行分析。在财务分析部分，需涵盖基本的财务关键指标。
+在宏观经济分析部分，考虑宏观经济指标对行业的影响，对 A 股新能源汽车行业整体发展趋势进行总结，并基于财务和宏观经济分析，对 TOP5 新能源车企的未来发展潜力和竞争态势做出比较和预测，指出各车企的优势与挑战。请以专业、严谨的语言，结合具体数据进行分析阐述。
+"""
 
 _, _, consumed_token = query(question, max_iter=1)
 print(f"Consumed tokens: {consumed_token}")