├── .env_example ├── data └── attention_is_all_you_need.pdf ├── utils.py ├── config.py ├── LICENSE ├── README.md ├── tools.py ├── app.py ├── agents.py ├── .gitignore ├── tasks.py └── requirements.txt /.env_example: -------------------------------------------------------------------------------- 1 | TAVILY_API_KEY = 2 | OPEN_ROUTER_API_KEY = 3 | GROQ_API_KEY = -------------------------------------------------------------------------------- /data/attention_is_all_you_need.pdf: -------------------------------------------------------------------------------- https://raw.githubusercontent.com/youssefHosni/Agentic-RAG-Application-DeepSeek/HEAD/data/attention_is_all_you_need.pdf -------------------------------------------------------------------------------- /utils.py: -------------------------------------------------------------------------------- 1 | import requests 2 | 3 | 4 | def download_pdf(url, filename): 5 | response = requests.get(url) 6 | with open(filename, 'wb') as file: 7 | file.write(response.content) 8 | 9 | -------------------------------------------------------------------------------- /config.py: -------------------------------------------------------------------------------- 1 | import os 2 | from dotenv import load_dotenv 3 | 4 | 5 | TAVILY_API_KEY = os.getenv("TAVILY_API_KEY") 6 | OPEN_ROUTER_API_KEY = os.getenv("OPEN_ROUTER_API_KEY") 7 | GROQ_API_KEY = os.getenv("GROQ_API_KEY") -------------------------------------------------------------------------------- /LICENSE: -------------------------------------------------------------------------------- 1 | MIT License 2 | 3 | Copyright (c) 2025 Youssef Hosni 4 | 5 | Permission is hereby granted, free of charge, to any person obtaining a copy 6 | of this software and associated documentation files (the "Software"), to deal 7 | in the Software without restriction, including without limitation the rights 8 | to use, copy, modify, merge, publish, distribute, sublicense, and/or sell 9 | copies of the Software, and to permit persons to whom the Software is 10 | furnished to do so, subject to the following conditions: 11 | 12 | The above copyright notice and this permission notice shall be included in all 13 | copies or substantial portions of the Software. 14 | 15 | THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR 16 | IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY, 17 | FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE 18 | AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER 19 | LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM, 20 | OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE 21 | SOFTWARE. 22 | -------------------------------------------------------------------------------- /README.md: -------------------------------------------------------------------------------- 1 | # Agentic-RAG-Application-DeepSeek 2 | 3 | ## Overview 4 | Agentic RAG is a project focused on developing a Retrieval-Augmented Generation (RAG) system with agentic capabilities. This system aims to enhance the retrieval and generation of information by incorporating intelligent agent behaviors. 5 | 6 | ![image](https://github.com/user-attachments/assets/58e29927-f186-4cc5-957d-05a67e9ab2b0) 7 | 8 | 9 | ## Features 10 | - **Intelligent Retrieval**: Efficiently retrieves relevant information from large datasets. 11 | - **Enhanced Generation**: Generates coherent and contextually appropriate responses. 12 | - **Agentic Capabilities**: Incorporates agent behaviors to improve interaction and decision-making. 13 | 14 | ## Installation 15 | To install the necessary dependencies, run: 16 | ```bash 17 | pip install -r requirements.txt 18 | ``` 19 | 20 | ## Usage 21 | To use the Agentic RAG system, follow these steps: 22 | 1. Clone the repository: 23 | ```bash 24 | git clone https://github.com/youssefhosni/agentic-rag.git 25 | ``` 26 | 2. Navigate to the project directory: 27 | ```bash 28 | cd agentic-rag 29 | ``` 30 | 3. Run the main script: 31 | ```bash 32 | python main.py 33 | ``` 34 | 35 | ## License 36 | This project is licensed under the MIT License. See the [LICENSE](LICENSE) file for details. 37 | 38 | ## Contact 39 | For any questions or inquiries, please contact [your email](mailto:Youssef.Hosni@outlook.com). 40 | -------------------------------------------------------------------------------- /tools.py: -------------------------------------------------------------------------------- 1 | from crewai_tools import PDFSearchTool 2 | from crewai.tools import tool 3 | from langchain_community.tools.tavily_search import TavilySearchResults 4 | from config import GROQ_API_KEY 5 | 6 | def create_pdf_tool(pdf_path): 7 | """ 8 | A tool to create a PDF search utility. 9 | 10 | Args: 11 | pdf_path (str): The path to the PDF file. 12 | 13 | Returns: 14 | PDFSearchTool: A configured PDF search tool instance. 15 | """ 16 | return PDFSearchTool( 17 | pdf=pdf_path, 18 | config=dict( 19 | llm=dict( 20 | provider="groq", 21 | config=dict( 22 | model="deepseek-r1-distill-qwen-32b", 23 | api_key=GROQ_API_KEY, 24 | # temperature=0.3, 25 | # max_tokens=2048, 26 | ), 27 | ), 28 | embedder=dict( 29 | provider="huggingface", 30 | config=dict( 31 | model="BAAI/bge-small-en-v1.5", 32 | ), 33 | ), 34 | ), 35 | ) 36 | 37 | 38 | @tool 39 | def web_search_tool(query): 40 | """ 41 | Web Search Tool. 42 | 43 | Args: 44 | query (str): The search query. 45 | 46 | Returns: 47 | str: The search results as text. 48 | """ 49 | web_search_tool = TavilySearchResults(k=3) 50 | return web_search_tool.run(query) 51 | -------------------------------------------------------------------------------- /app.py: -------------------------------------------------------------------------------- 1 | import time 2 | from crewai import Crew 3 | from tools import create_pdf_tool, web_search_tool 4 | from agents import create_router_agent, create_retriever_agent, create_grader_agent, create_hallucination_grader, create_answer_grader 5 | from tasks import create_router_task, create_retriever_task, create_grader_task, create_hallucination_task, create_answer_task 6 | 7 | def main(): 8 | # Download the PDF 9 | pdf_filename = 'data/attention_is_all_you_need.pdf' 10 | 11 | # Create tools 12 | rag_tool = create_pdf_tool(pdf_filename) 13 | 14 | # Create agents 15 | router_agent = create_router_agent() 16 | retriever_agent = create_retriever_agent() 17 | grader_agent = create_grader_agent() 18 | hallucination_grader = create_hallucination_grader() 19 | answer_grader = create_answer_grader() 20 | 21 | # Create tasks 22 | router_task = create_router_task(router_agent) 23 | retriever_task = create_retriever_task(retriever_agent, router_task, rag_tool, web_search_tool) 24 | grader_task = create_grader_task(grader_agent, retriever_task) 25 | hallucination_task = create_hallucination_task(hallucination_grader, grader_task) 26 | answer_task = create_answer_task(answer_grader, hallucination_task, web_search_tool) 27 | 28 | # Create crew 29 | rag_crew = Crew( 30 | agents=[router_agent, retriever_agent, grader_agent, hallucination_grader, answer_grader], 31 | tasks=[router_task, retriever_task, grader_task, hallucination_task, answer_task], 32 | verbose=True, 33 | ) 34 | start_time = time.time() 35 | # Kickoff the crew with a user question 36 | user_question = {"question": "what is the weather in New York city?"} 37 | result = rag_crew.kickoff(inputs=user_question) 38 | end_time = time.time() 39 | print(f"Time taken: {end_time - start_time} seconds.") 40 | print(result) 41 | 42 | 43 | if __name__ == "__main__": 44 | main() -------------------------------------------------------------------------------- /agents.py: -------------------------------------------------------------------------------- 1 | from crewai import Agent 2 | from crewai import LLM 3 | from config import OPEN_ROUTER_API_KEY 4 | 5 | #llm = LLM(model="deepseek-r1:1.5b", base_url="http://localhost:11434") 6 | 7 | llm = LLM(model="openrouter/deepseek/deepseek-r1", 8 | temperature=0, 9 | api_key=OPEN_ROUTER_API_KEY 10 | ) 11 | 12 | def create_router_agent(): 13 | return Agent( 14 | role='Router', 15 | goal='Route user questions to either vectorstore or web search based on content relevance', 16 | backstory=( 17 | "You are an expert at determining whether a question can be answered using the " 18 | "information stored in our vector database, or requires a web search. " 19 | "You understand that the vector database contains comprehensive knowledge base " 20 | "You make routing decisions based on the semantic meaning of questions rather than just keyword matching." 21 | ), 22 | verbose=True, 23 | allow_delegation=False, 24 | llm=llm, 25 | ) 26 | 27 | 28 | def create_retriever_agent(): 29 | return Agent( 30 | role="Retriever", 31 | goal="Use the information retrieved from the vectorstore to answer the question", 32 | backstory=( 33 | "You are an assistant for question-answering tasks." 34 | "Use the information present in the retrieved context to answer the question." 35 | "You have to provide a clear concise answer." 36 | ), 37 | verbose=True, 38 | allow_delegation=False, 39 | llm=llm, 40 | ) 41 | 42 | 43 | def create_grader_agent(): 44 | return Agent( 45 | role='Answer Grader', 46 | goal='Filter out erroneous retrievals', 47 | backstory=( 48 | "You are a grader assessing relevance of a retrieved document to a user question." 49 | "If the document contains keywords related to the user question, grade it as relevant." 50 | "It does not need to be a stringent test.You have to make sure that the answer is relevant to the question." 51 | ), 52 | verbose=True, 53 | allow_delegation=False, 54 | llm=llm, 55 | ) 56 | 57 | 58 | def create_hallucination_grader(): 59 | return Agent( 60 | role="Hallucination Grader", 61 | goal="Filter out hallucination", 62 | backstory=( 63 | "You are a hallucination grader assessing whether an answer is grounded in / supported by a set of facts." 64 | "Make sure you meticulously review the answer and check if the response provided is in alignmnet with the question asked" 65 | ), 66 | verbose=True, 67 | allow_delegation=False, 68 | llm=llm, 69 | ) 70 | 71 | 72 | def create_answer_grader(): 73 | return Agent( 74 | role="Answer Grader", 75 | goal="Filter out hallucination from the answer.", 76 | backstory=( 77 | "You are a grader assessing whether an answer is useful to resolve a question." 78 | "Make sure you meticulously review the answer and check if it makes sense for the question asked" 79 | "If the answer is relevant generate a clear and concise response." 80 | "If the answer gnerated is not relevant then perform a websearch using 'web_search_tool'" 81 | ), 82 | verbose=True, 83 | allow_delegation=False, 84 | llm=llm, 85 | ) 86 | -------------------------------------------------------------------------------- /.gitignore: -------------------------------------------------------------------------------- 1 | # Byte-compiled / optimized / DLL files 2 | __pycache__/ 3 | *.py[cod] 4 | *$py.class 5 | 6 | # C extensions 7 | *.so 8 | 9 | # Distribution / packaging 10 | .Python 11 | build/ 12 | develop-eggs/ 13 | dist/ 14 | downloads/ 15 | eggs/ 16 | .eggs/ 17 | lib/ 18 | lib64/ 19 | parts/ 20 | sdist/ 21 | var/ 22 | wheels/ 23 | share/python-wheels/ 24 | *.egg-info/ 25 | .installed.cfg 26 | *.egg 27 | MANIFEST 28 | 29 | # PyInstaller 30 | # Usually these files are written by a python script from a template 31 | # before PyInstaller builds the exe, so as to inject date/other infos into it. 32 | *.manifest 33 | *.spec 34 | 35 | # Installer logs 36 | pip-log.txt 37 | pip-delete-this-directory.txt 38 | 39 | # Unit test / coverage reports 40 | htmlcov/ 41 | .tox/ 42 | .nox/ 43 | .coverage 44 | .coverage.* 45 | .cache 46 | nosetests.xml 47 | coverage.xml 48 | *.cover 49 | *.py,cover 50 | .hypothesis/ 51 | .pytest_cache/ 52 | cover/ 53 | 54 | # Translations 55 | *.mo 56 | *.pot 57 | 58 | # Django stuff: 59 | *.log 60 | local_settings.py 61 | db.sqlite3 62 | db.sqlite3-journal 63 | 64 | # Flask stuff: 65 | instance/ 66 | .webassets-cache 67 | 68 | # Scrapy stuff: 69 | .scrapy 70 | 71 | # Sphinx documentation 72 | docs/_build/ 73 | 74 | # PyBuilder 75 | .pybuilder/ 76 | target/ 77 | 78 | # Jupyter Notebook 79 | .ipynb_checkpoints 80 | 81 | # IPython 82 | profile_default/ 83 | ipython_config.py 84 | 85 | # pyenv 86 | # For a library or package, you might want to ignore these files since the code is 87 | # intended to run in multiple environments; otherwise, check them in: 88 | # .python-version 89 | 90 | # pipenv 91 | # According to pypa/pipenv#598, it is recommended to include Pipfile.lock in version control. 92 | # However, in case of collaboration, if having platform-specific dependencies or dependencies 93 | # having no cross-platform support, pipenv may install dependencies that don't work, or not 94 | # install all needed dependencies. 95 | #Pipfile.lock 96 | 97 | # UV 98 | # Similar to Pipfile.lock, it is generally recommended to include uv.lock in version control. 99 | # This is especially recommended for binary packages to ensure reproducibility, and is more 100 | # commonly ignored for libraries. 101 | #uv.lock 102 | 103 | # poetry 104 | # Similar to Pipfile.lock, it is generally recommended to include poetry.lock in version control. 105 | # This is especially recommended for binary packages to ensure reproducibility, and is more 106 | # commonly ignored for libraries. 107 | # https://python-poetry.org/docs/basic-usage/#commit-your-poetrylock-file-to-version-control 108 | #poetry.lock 109 | 110 | # pdm 111 | # Similar to Pipfile.lock, it is generally recommended to include pdm.lock in version control. 112 | #pdm.lock 113 | # pdm stores project-wide configurations in .pdm.toml, but it is recommended to not include it 114 | # in version control. 115 | # https://pdm.fming.dev/latest/usage/project/#working-with-version-control 116 | .pdm.toml 117 | .pdm-python 118 | .pdm-build/ 119 | 120 | # PEP 582; used by e.g. github.com/David-OConnor/pyflow and github.com/pdm-project/pdm 121 | __pypackages__/ 122 | 123 | # Celery stuff 124 | celerybeat-schedule 125 | celerybeat.pid 126 | 127 | # SageMath parsed files 128 | *.sage.py 129 | 130 | # Environments 131 | .env 132 | .venv 133 | env/ 134 | venv/ 135 | ENV/ 136 | env.bak/ 137 | venv.bak/ 138 | 139 | # Spyder project settings 140 | .spyderproject 141 | .spyproject 142 | 143 | # Rope project settings 144 | .ropeproject 145 | 146 | # mkdocs documentation 147 | /site 148 | 149 | # mypy 150 | .mypy_cache/ 151 | .dmypy.json 152 | dmypy.json 153 | 154 | # Pyre type checker 155 | .pyre/ 156 | 157 | # pytype static type analyzer 158 | .pytype/ 159 | 160 | # Cython debug symbols 161 | cython_debug/ 162 | 163 | # PyCharm 164 | # JetBrains specific template is maintained in a separate JetBrains.gitignore that can 165 | # be found at https://github.com/github/gitignore/blob/main/Global/JetBrains.gitignore 166 | # and can be added to the global gitignore or merged into this file. For a more nuclear 167 | # option (not recommended) you can uncomment the following to ignore the entire idea folder. 168 | #.idea/ 169 | 170 | # PyPI configuration file 171 | .pypirc 172 | -------------------------------------------------------------------------------- /tasks.py: -------------------------------------------------------------------------------- 1 | from crewai import Task 2 | 3 | def create_router_task(router_agent): 4 | return Task( 5 | description=( 6 | "Analyze the given question {question} to determine the appropriate search method:\n" 7 | "\n" 8 | "1. Use 'vectorstore' if:\n" 9 | " - The question contains a keyword or a similar words\n" 10 | " - The topic is likely covered in our vector database\n" 11 | "\n" 12 | "2. Use 'web_search' if:\n" 13 | " - The topic requires current or real-time information\n" 14 | " - The question is about general topics not covered in our vector database\n" 15 | "\n" 16 | "Make decisions based on semantic understanding rather than keyword matching." 17 | ), 18 | expected_output=( 19 | "Return exactly one word:\n" 20 | "'vectorstore' - if the question can be answered from our RAG knowledge base\n" 21 | "'web_search' - if the question requires external information\n" 22 | "No additional explanation or preamble should be included." 23 | ), 24 | agent=router_agent, 25 | ) 26 | 27 | 28 | def create_retriever_task(retriever_agent, router_task, rag_tool, web_search_tool): 29 | return Task( 30 | description=( 31 | "Based on the response from the router task extract information for the question {question} with the help of the respective tool." 32 | "Use the web_serach_tool to retrieve information from the web in case the router task output is 'websearch'. You should pass the input query {question} to the web_search_tool." 33 | "Use the rag_tool to retrieve information from the vectorstore in case the router task output is 'vectorstore'." 34 | ), 35 | expected_output=( 36 | "You should analyse the output of the 'router_task'" 37 | "If the response is 'websearch' then use the web_search_tool to retrieve information from the web." 38 | "If the response is 'vectorstore' then use the rag_tool to retrieve information from the vectorstore." 39 | "Return a claer and consise text as response." 40 | ), 41 | agent=retriever_agent, 42 | context=[router_task], 43 | tools=[rag_tool, web_search_tool], 44 | ) 45 | 46 | 47 | def create_grader_task(grader_agent, retriever_task): 48 | return Task( 49 | description=( 50 | "Based on the response from the retriever task for the quetion {question} evaluate whether the retrieved content is relevant to the question." 51 | ), 52 | expected_output=( 53 | "Binary score 'yes' or 'no' score to indicate whether the document is relevant to the question" 54 | "You must answer 'yes' if the response from the 'retriever_task' is in alignment with the question asked." 55 | "You must answer 'no' if the response from the 'retriever_task' is not in alignment with the question asked." 56 | "Do not provide any preamble or explanations except for 'yes' or 'no'." 57 | ), 58 | agent=grader_agent, 59 | context=[retriever_task], 60 | ) 61 | 62 | 63 | def create_hallucination_task(hallucination_grader, grader_task): 64 | return Task( 65 | description=( 66 | "Based on the response from the grader task for the quetion {question} evaluate whether the answer is grounded in / supported by a set of facts." 67 | ), 68 | expected_output=( 69 | "Binary score 'yes' or 'no' score to indicate whether the answer is sync with the question asked" 70 | "Respond 'yes' if the answer is in useful and contains fact about the question asked." 71 | "Respond 'no' if the answer is not useful and does not contains fact about the question asked." 72 | "Do not provide any preamble or explanations except for 'yes' or 'no'." 73 | ), 74 | agent=hallucination_grader, 75 | context=[grader_task], 76 | ) 77 | 78 | 79 | def create_answer_task(answer_grader, hallucination_task, web_search_tool): 80 | return Task( 81 | description=( 82 | "Based on the response from the hallucination task for the quetion {question} evaluate whether the answer is useful to resolve the question." 83 | "If the answer is 'yes' return a clear and concise answer." 84 | "If the answer is 'no' then perform a 'websearch' and return the response" 85 | ), 86 | expected_output=( 87 | "Return a clear and concise response if the response from 'hallucination_task' is 'yes'." 88 | "Perform a web search using 'web_search_tool' and return ta clear and concise response only if the response from 'hallucination_task' is 'no'." 89 | "Otherwise respond as 'Sorry! unable to find a valid response'." 90 | "Make sure the final response is clear and concise and contain only the answer to the input question without any preamble or explanation as this answer will be presnted to the user." 91 | "The final answer should be a clear and concise response to the input question." 92 | ), 93 | context=[hallucination_task], 94 | agent=answer_grader, 95 | tool = [web_search_tool], 96 | 97 | ) -------------------------------------------------------------------------------- /requirements.txt: -------------------------------------------------------------------------------- 1 | aiohappyeyeballs==2.4.6 2 | aiohttp==3.11.12 3 | aiosignal==1.3.2 4 | alembic==1.14.1 5 | annotated-types==0.7.0 6 | anyio==4.8.0 7 | appdirs==1.4.4 8 | asgiref==3.8.1 9 | asttokens==3.0.0 10 | attrs==25.1.0 11 | auth0-python==4.8.0 12 | backoff==2.2.1 13 | bcrypt==4.2.1 14 | beautifulsoup4==4.13.3 15 | blinker==1.9.0 16 | build==1.2.2.post1 17 | cachetools==5.5.1 18 | certifi==2025.1.31 19 | cffi==1.17.1 20 | charset-normalizer==3.4.1 21 | chroma-hnswlib==0.7.6 22 | chromadb==0.5.23 23 | click==8.1.8 24 | cohere==5.13.12 25 | colorama==0.4.6 26 | coloredlogs==15.0.1 27 | crewai==0.102.0 28 | crewai-tools==0.36.0 29 | cryptography==44.0.1 30 | dataclasses-json==0.6.7 31 | decorator==5.1.1 32 | Deprecated==1.2.18 33 | deprecation==2.1.0 34 | distro==1.9.0 35 | docker==7.1.0 36 | docstring_parser==0.16 37 | durationpy==0.9 38 | embedchain==0.1.126 39 | et_xmlfile==2.0.0 40 | executing==2.2.0 41 | fastapi==0.115.8 42 | fastavro==1.10.0 43 | filelock==3.17.0 44 | flatbuffers==25.2.10 45 | frozenlist==1.5.0 46 | fsspec==2025.2.0 47 | google-api-core==2.24.1 48 | google-auth==2.38.0 49 | google-cloud-aiplatform==1.80.0 50 | google-cloud-bigquery==3.29.0 51 | google-cloud-core==2.4.1 52 | google-cloud-resource-manager==1.14.0 53 | google-cloud-storage==2.19.0 54 | google-crc32c==1.6.0 55 | google-resumable-media==2.7.2 56 | googleapis-common-protos==1.67.0 57 | gptcache==0.1.44 58 | greenlet==3.1.1 59 | groq==0.18.0 60 | grpc-google-iam-v1==0.14.0 61 | grpcio==1.70.0 62 | grpcio-status==1.70.0 63 | grpcio-tools==1.70.0 64 | h11==0.14.0 65 | h2==4.2.0 66 | hpack==4.1.0 67 | httpcore==1.0.7 68 | httptools==0.6.4 69 | httpx==0.27.2 70 | httpx-sse==0.4.0 71 | huggingface-hub==0.28.1 72 | humanfriendly==10.0 73 | hyperframe==6.1.0 74 | idna==3.10 75 | importlib_metadata==8.5.0 76 | importlib_resources==6.5.2 77 | instructor==1.7.2 78 | ipython==8.32.0 79 | jedi==0.19.2 80 | Jinja2==3.1.5 81 | jiter==0.8.2 82 | joblib==1.4.2 83 | json5==0.10.0 84 | json_repair==0.36.1 85 | jsonpatch==1.33 86 | jsonpickle==4.0.1 87 | jsonpointer==3.0.0 88 | jsonref==1.1.0 89 | jsonschema==4.23.0 90 | jsonschema-specifications==2024.10.1 91 | kubernetes==32.0.0 92 | lancedb==0.19.0 93 | langchain==0.3.18 94 | langchain-cohere==0.3.5 95 | langchain-community==0.3.17 96 | langchain-core==0.3.35 97 | langchain-experimental==0.3.4 98 | langchain-groq==0.2.4 99 | langchain-huggingface==0.1.2 100 | langchain-openai==0.2.14 101 | langchain-text-splitters==0.3.6 102 | langsmith==0.1.147 103 | litellm==1.60.2 104 | Mako==1.3.9 105 | markdown-it-py==3.0.0 106 | MarkupSafe==3.0.2 107 | marshmallow==3.26.1 108 | matplotlib-inline==0.1.7 109 | mdurl==0.1.2 110 | mem0ai==0.1.48 111 | mmh3==5.1.0 112 | monotonic==1.6 113 | mpmath==1.3.0 114 | multidict==6.1.0 115 | mypy-extensions==1.0.0 116 | networkx==3.4.2 117 | nodeenv==1.9.1 118 | numpy==1.26.4 119 | oauthlib==3.2.2 120 | onnxruntime==1.20.1 121 | openai==1.62.0 122 | openpyxl==3.1.5 123 | opentelemetry-api==1.30.0 124 | opentelemetry-exporter-otlp-proto-common==1.30.0 125 | opentelemetry-exporter-otlp-proto-grpc==1.30.0 126 | opentelemetry-exporter-otlp-proto-http==1.30.0 127 | opentelemetry-instrumentation==0.51b0 128 | opentelemetry-instrumentation-asgi==0.51b0 129 | opentelemetry-instrumentation-fastapi==0.51b0 130 | opentelemetry-proto==1.30.0 131 | opentelemetry-sdk==1.30.0 132 | opentelemetry-semantic-conventions==0.51b0 133 | opentelemetry-util-http==0.51b0 134 | orjson==3.10.15 135 | overrides==7.7.0 136 | packaging==24.2 137 | pandas==2.2.3 138 | parso==0.8.4 139 | pdfminer.six==20231228 140 | pdfplumber==0.11.5 141 | pillow==11.1.0 142 | portalocker==2.10.1 143 | posthog==3.13.0 144 | prompt_toolkit==3.0.50 145 | propcache==0.2.1 146 | proto-plus==1.26.0 147 | protobuf==5.29.3 148 | pure_eval==0.2.3 149 | pyarrow==19.0.0 150 | pyasn1==0.6.1 151 | pyasn1_modules==0.4.1 152 | pycparser==2.22 153 | pydantic==2.10.6 154 | pydantic-settings==2.7.1 155 | pydantic_core==2.27.2 156 | Pygments==2.19.1 157 | PyJWT==2.10.1 158 | pylance==0.23.0 159 | pypdf==5.3.0 160 | pypdfium2==4.30.1 161 | PyPika==0.48.9 162 | pyproject_hooks==1.2.0 163 | pyreadline3==3.5.4 164 | pyright==1.1.394 165 | pysbd==0.3.4 166 | python-dateutil==2.9.0.post0 167 | python-dotenv==1.0.1 168 | pytube==15.0.0 169 | pytz==2024.2 170 | pyvis==0.3.2 171 | pywin32==308 172 | PyYAML==6.0.2 173 | qdrant-client==1.13.2 174 | referencing==0.36.2 175 | regex==2024.11.6 176 | requests==2.32.3 177 | requests-oauthlib==2.0.0 178 | requests-toolbelt==1.0.0 179 | rich==13.9.4 180 | rpds-py==0.22.3 181 | rsa==4.9 182 | safetensors==0.5.2 183 | schema==0.7.7 184 | scikit-learn==1.6.1 185 | scipy==1.15.1 186 | sentence-transformers==3.4.1 187 | shapely==2.0.7 188 | shellingham==1.5.4 189 | six==1.17.0 190 | sniffio==1.3.1 191 | soupsieve==2.6 192 | SQLAlchemy==2.0.38 193 | stack-data==0.6.3 194 | starlette==0.45.3 195 | sympy==1.13.1 196 | tabulate==0.9.0 197 | tenacity==9.0.0 198 | threadpoolctl==3.5.0 199 | tiktoken==0.7.0 200 | tokenizers==0.21.0 201 | tomli==2.2.1 202 | tomli_w==1.2.0 203 | torch==2.6.0 204 | tqdm==4.67.1 205 | traitlets==5.14.3 206 | transformers==4.48.3 207 | typer==0.15.1 208 | types-requests==2.32.0.20241016 209 | typing-inspect==0.9.0 210 | typing_extensions==4.12.2 211 | tzdata==2025.1 212 | urllib3==2.3.0 213 | uv==0.5.31 214 | uvicorn==0.34.0 215 | watchfiles==1.0.4 216 | wcwidth==0.2.13 217 | websocket-client==1.8.0 218 | websockets==14.2 219 | wrapt==1.17.2 220 | yarl==1.18.3 221 | zipp==3.21.0 222 | --------------------------------------------------------------------------------