Merge branch 'main' into docker_en

README.md

@@ -64,7 +64,7 @@ Our vision is to revolutionize how AI agents collaborate to solve real-world tas
 - [📋 Table of Contents](#-table-of-contents)
 - [🔥 News](#-news)
 - [🎬 Demo Video](#-demo-video)
-- [✨️ Core Features](#-code-features)
+- [✨️ Core Features](#-core-features)
 - [🛠️ Installation](#️-installation)
   - [**Clone the Github repository**](#clone-the-github-repository)
   - [**Set up Environment**](#set-up-environment)
@@ -139,7 +139,10 @@ playwright install
 In the `owl/.env_template` file, you will find all the necessary API keys along with the websites where you can register for each service. To use these API services, follow these steps:
 
 1. *Copy and Rename*: Duplicate the `.env_example` file and rename the copy to `.env`.
-2. *Fill in Your Keys*: Open the `.env` file and insert your API keys in the corresponding fields. 
+```bash
+cp owl/.env_template .env
+```
+2. *Fill in Your Keys*: Open the `.env` file and insert your API keys in the corresponding fields.  (For the minimal example (`run_mini.py`), you only need to configure the LLM API key (e.g., OPENAI_API_KEY).)
 3. *For using more other models*: please refer to our CAMEL models docs:https://docs.camel-ai.org/key_modules/models.html#supported-model-platforms-in-camel
 
 
@@ -171,11 +174,18 @@ For more detailed Docker usage instructions, including cross-platform support, o
 
 
    
-Run the following minimal example:
+Run the following demo case:
 
 ```bash
 python owl/run.py
 ```
+
+For a simpler version that only requires an LLM API key, you can try our minimal example:
+
+```bash
+python owl/run_mini.py
+```
+
 You can run OWL agent with your own task by modifying the `run.py` script:
 
 ```python
@@ -188,6 +198,21 @@ answer, chat_history, token_count = run_society(society)
 logger.success(f"Answer: {answer}")
 ```
 
+For uploading files, simply provide the file path along with your question:
+
+```python
+# Task with a local file (e.g., file path: `tmp/example.docx`)
+question = "What is in the given DOCX file? Here is the file path: tmp/example.docx"
+
+society = construct_society(question)
+answer, chat_history, token_count = run_society(society)
+
+logger.success(f"Answer: {answer}")
+```
+
+OWL will then automatically invoke document-related tools to process the file and extract the answer.
+
+
 Example tasks you can try:
 - "Find the latest stock price for Apple Inc."
 - "Analyze the sentiment of recent tweets about climate change"

README_zh.md

@@ -165,13 +165,19 @@ docker-compose exec owl bash -c "xvfb-python run.py"
 
 # 🚀 快速开始
    
-运行以下最小示例：
+运行以下示例：
 
 ```bash
 python owl/run.py
 ```
 
-你可以通过修改 `run.py` 来运行自定义任务的 OWL 智能体：
+我们还提供了一个最小化示例，只需配置LLM的API密钥即可运行：
+
+```bash
+python owl/run_mini.py
+```
+
+你可以通过修改 `run.py` 脚本来运行自己的任务：
 
 ```python
 # Define your own task
@@ -183,11 +189,29 @@ answer, chat_history, token_count = run_society(society)
 logger.success(f"Answer: {answer}")
 ```
 
+上传文件时，只需提供文件路径和问题：
+
+```python
+# 处理本地文件（例如，文件路径为 `tmp/example.docx`）
+question = "给定的 DOCX 文件中有什么内容？文件路径如下：tmp/example.docx"
+
+society = construct_society(question)
+answer, chat_history, token_count = run_society(society)
+
+logger.success(f"答案：{answer}")
+```
+
+OWL 将自动调用与文档相关的工具来处理文件并提取答案。
+
+
+OWL 将自动调用与文档相关的工具来处理文件并提取答案。
+
 你可以尝试以下示例任务：
 - "查询苹果公司的最新股票价格"
 - "分析关于气候变化的最新推文情绪"
 - "帮我调试这段 Python 代码：[在此粘贴你的代码]"
 - "总结这篇研究论文的主要观点：[论文URL]"
+- 
 # 🧪 实验
 
 我们提供了一个脚本用于复现 GAIA 上的实验结果。  

owl/run.py

@@ -2,17 +2,23 @@ from dotenv import load_dotenv
 load_dotenv()
 
 from camel.models import ModelFactory
-from camel.toolkits import *
+from camel.toolkits import (
+    WebToolkit, 
+    DocumentProcessingToolkit, 
+    VideoAnalysisToolkit, 
+    AudioAnalysisToolkit, 
+    CodeExecutionToolkit, 
+    ImageAnalysisToolkit, 
+    SearchToolkit,
+    ExcelToolkit
+    )
 from camel.types import ModelPlatformType, ModelType
-from camel.configs import ChatGPTConfig
+# from camel.configs import ChatGPTConfig
 
-from typing import List, Dict
 
-from retry import retry
 from loguru import logger
 
 from utils import OwlRolePlaying, run_society
-import os
 
 
 
@@ -25,13 +31,13 @@ def construct_society(question: str) -> OwlRolePlaying:
     user_model = ModelFactory.create(
         model_platform=ModelPlatformType.OPENAI,
         model_type=ModelType.GPT_4O,
-        model_config_dict=ChatGPTConfig(temperature=0, top_p=1).as_dict(), # [Optional] the config for model
+        # model_config_dict=ChatGPTConfig(temperature=0, top_p=1).as_dict(), # [Optional] the config for model
     )
 
     assistant_model = ModelFactory.create(
         model_platform=ModelPlatformType.OPENAI,
         model_type=ModelType.GPT_4O,
-        model_config_dict=ChatGPTConfig(temperature=0, top_p=1).as_dict(), # [Optional] the config for model
+        # model_config_dict=ChatGPTConfig(temperature=0, top_p=1).as_dict(), # [Optional] the config for model
     )
 
     tools_list = [

owl/run_deepseek_example.py

@@ -3,13 +3,10 @@ from camel.toolkits import *
 from camel.types import ModelPlatformType, ModelType
 from camel.configs import DeepSeekConfig
 
-from typing import List, Dict
 from dotenv import load_dotenv
-from retry import retry
 from loguru import logger
 
 from utils import OwlRolePlaying, run_society
-import os
 
 
 load_dotenv()

owl/run_gaia_roleplaying.py

@@ -5,11 +5,9 @@ from camel.configs import ChatGPTConfig
 from utils import GAIABenchmark
 
 from dotenv import load_dotenv
-from retry import retry
 from loguru import logger
 
 import os
-import shutil
 
 load_dotenv()
 

owl/run_mini.py

@@ -0,0 +1,77 @@
+from dotenv import load_dotenv
+load_dotenv()
+
+from camel.models import ModelFactory
+from camel.toolkits import (
+    WebToolkit, 
+    SearchToolkit,
+    FunctionTool
+    )
+from camel.types import ModelPlatformType, ModelType
+
+
+from loguru import logger
+
+from utils import OwlRolePlaying, run_society
+
+
+
+def construct_society(question: str) -> OwlRolePlaying:
+    r"""Construct the society based on the question."""
+
+    user_role_name = "user"
+    assistant_role_name = "assistant"
+    
+    user_model = ModelFactory.create(
+        model_platform=ModelPlatformType.OPENAI,
+        model_type=ModelType.GPT_4O,
+    )
+
+    assistant_model = ModelFactory.create(
+        model_platform=ModelPlatformType.OPENAI,
+        model_type=ModelType.GPT_4O,
+    )
+
+    tools_list = [
+        *WebToolkit(
+            headless=False, 
+            web_agent_model=assistant_model, 
+            planning_agent_model=assistant_model
+        ).get_tools(),
+        FunctionTool(SearchToolkit(model=assistant_model).search_duckduckgo),
+    ]
+
+    user_role_name = 'user'
+    user_agent_kwargs = dict(model=user_model)
+    assistant_role_name = 'assistant'
+    assistant_agent_kwargs = dict(model=assistant_model,
+    tools=tools_list)
+    
+    task_kwargs = {
+        'task_prompt': question,
+        'with_task_specify': False,
+    }
+
+    society = OwlRolePlaying(
+        **task_kwargs,
+        user_role_name=user_role_name,
+        user_agent_kwargs=user_agent_kwargs,
+        assistant_role_name=assistant_role_name,
+        assistant_agent_kwargs=assistant_agent_kwargs,
+    )
+    
+    return society
+
+
+# Example case
+question = "What was the volume in m^3 of the fish bag that was calculated in the University of Leicester paper `Can Hiccup Supply Enough Fish to Maintain a Dragon’s Diet?` "
+
+society = construct_society(question)
+answer, chat_history, token_count = run_society(society)
+
+logger.success(f"Answer: {answer}")
+
+
+
+
+

owl/{run_qwq_demo.py → run_openai_compatiable_model.py}

@@ -26,7 +26,7 @@ def construct_society(question: str) -> OwlRolePlaying:
     
     user_model = ModelFactory.create(
         model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL,
-        model_type="qwq-32b",
+        model_type="qwen-max",
         api_key=os.getenv("QWEN_API_KEY"),
         url="https://dashscope.aliyuncs.com/compatible-mode/v1",
         model_config_dict={"temperature": 0.4, "max_tokens": 4096},
@@ -34,7 +34,7 @@ def construct_society(question: str) -> OwlRolePlaying:
 
     assistant_model = ModelFactory.create(
         model_platform=ModelPlatformType.OPENAI_COMPATIBLE_MODEL,
-        model_type="qwq-32b",
+        model_type="qwen-max",
         api_key=os.getenv("QWEN_API_KEY"),
         url="https://dashscope.aliyuncs.com/compatible-mode/v1",
         model_config_dict={"temperature": 0.4, "max_tokens": 4096},
@@ -79,7 +79,7 @@ def construct_society(question: str) -> OwlRolePlaying:
 
 
 # Example case
-question = "What was the volume in m^3 of the fish bag that was calculated in the University of Leicester paper `Can Hiccup Supply Enough Fish to Maintain a Dragon’s Diet?` "
+question = "what is the weather in beijing today?"
 
 society = construct_society(question)
 answer, chat_history, token_count = run_society(society)