> ## Documentation Index > Fetch the complete documentation index at: https://docs.ag2.ai/llms.txt > Use this file to discover all available pages before exploring further. # All Notebooks export const notebooksMetadata = [{ "title": "Websockets: Streaming input and output using websockets", "link": "/docs/use-cases/notebooks/notebooks/agentchat_websockets", "description": "Websockets facilitate real-time, bidirectional communication between web clients and servers, enhancing the responsiveness and interactivity of AG2-powered applications.", "image": null, "tags": ["websockets", "streaming"], "source": "/notebook/agentchat_websockets.ipynb" }, { "title": "Perplexity Search Tool", "link": "/docs/use-cases/notebooks/notebooks/tools_perplexity_search", "description": "Perplexity Search Integration with AG2", "image": null, "tags": ["tools", "perplexity", "web-search", "search"], "source": "/notebook/tools_perplexity_search.ipynb" }, { "title": "Auto Generated Agent Chat: Task Solving with Code Generation, Execution, Debugging & Human Feedback", "link": "/docs/use-cases/notebooks/notebooks/agentchat_human_feedback", "description": "Code generation, execution, debugging and human feedback.", "image": null, "tags": ["human", "code generation"], "source": "/notebook/agentchat_human_feedback.ipynb" }, { "title": "Using a local Telemetry server to monitor a GraphRAG agent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_swarm_graphrag_telemetry_trip_planner", "description": "FalkorDB GraphRAG utilises a knowledge graph and can be added as a capability to agents. Together with a swarm orchestration of agents is highly effective at providing a RAG capability.", "image": null, "tags": ["RAG", "tool/function", "swarm"], "source": "/notebook/agentchat_swarm_graphrag_telemetry_trip_planner.ipynb" }, { "title": "Auto Generated Agent Chat: Task Solving with Provided Tools as Functions", "link": "/docs/use-cases/notebooks/notebooks/agentchat_function_call", "description": "Register function calls using AssistantAgent and UserProxyAgent to execute python or shell code in customized ways. Demonstrating two ways of registering functions.", "image": null, "tags": ["code generation", "tool/function"], "source": "/notebook/agentchat_function_call.ipynb" }, { "title": "RealtimeAgent with gemini client", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_gemini_websocket", "description": "RealtimeAgent with gemini client using websockets", "image": null, "tags": ["realtime", "websockets", "gemini"], "source": "/notebook/agentchat_realtime_gemini_websocket.ipynb" }, { "title": "Agentic RAG workflow on tabular data from a PDF file", "link": "/docs/use-cases/notebooks/notebooks/agentchat_tabular_data_rag_workflow", "description": "Agentic RAG workflow on tabular data from a PDF file", "image": null, "tags": ["RAG", "groupchat"], "source": "/notebook/agentchat_tabular_data_rag_workflow.ipynb" }, { "title": "Language Agent Tree Search", "link": "/docs/use-cases/notebooks/notebooks/lats_search", "description": "Language Agent Tree Search.", "image": null, "tags": ["LATS", "search", "reasoning", "reflection"], "source": "/notebook/lats_search.ipynb" }, { "title": "Config loader utility functions", "link": "/docs/use-cases/notebooks/notebooks/config_loader_utility_functions", "description": "Config loader utility functions", "image": null, "tags": ["utility", "config"], "source": "/notebook/config_loader_utility_functions.ipynb" }, { "title": "Wikipedia Agent", "link": "/docs/use-cases/notebooks/notebooks/agents_wikipedia", "description": "Search Wikipedia with WikipediaAgent", "image": null, "tags": ["tools", "wikipedia", "search"], "source": "/notebook/agents_wikipedia.ipynb" }, { "title": "Google Drive Tools", "link": "/docs/use-cases/notebooks/notebooks/tools_google_drive", "description": "Google Drive Tools", "image": null, "tags": ["agents", "tools", "google drive"], "source": "/notebook/tools_google_drive.ipynb" }, { "title": "FSM - User can input speaker transition constraints", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_finite_state_machine", "description": "Explore the demonstration of the Finite State Machine implementation, which allows the user to input speaker transition constraints.", "image": null, "tags": ["group chat", "fsm", "orchestration"], "source": "/notebook/agentchat_groupchat_finite_state_machine.ipynb" }, { "title": "DeepResearchAgent", "link": "/docs/use-cases/notebooks/notebooks/agents_deep_researcher", "description": "DeepResearch Agent", "image": null, "tags": ["agents", "browser-use", "DeepResearch", "webscraping", "function calling"], "source": "/notebook/agents_deep_researcher.ipynb" }, { "title": "Group Chat with Retrieval Augmented Generation", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_RAG", "description": "Implement and manage a multi-agent chat system using AG2, where AI assistants retrieve information, generate code, and interact collaboratively to solve complex tasks, especially in areas not covered by their training data.", "image": null, "tags": ["group chat", "orchestration", "RAG"], "source": "/notebook/agentchat_groupchat_RAG.ipynb" }, { "title": "AgentOptimizer: An Agentic Way to Train Your LLM Agent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_agentoptimizer", "description": "AgentOptimizer is able to prompt LLMs to iteratively optimize function/skills of AutoGen agents according to the historical conversation and performance.", "image": null, "tags": ["optimization", "tool/function"], "source": "/notebook/agentchat_agentoptimizer.ipynb" }, { "title": "Using RetrieveChat with Qdrant for Retrieve Augmented Code Generation and Question Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_RetrieveChat_qdrant", "description": "This notebook demonstrates the usage of QdrantRetrieveUserProxyAgent for RAG.", "image": null, "tags": ["Qdrant", "integration", "RAG"], "source": "/notebook/agentchat_RetrieveChat_qdrant.ipynb" }, { "title": "RealtimeAgent with WebRTC connection", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_webrtc", "description": "RealtimeAgent using websockets", "image": null, "tags": ["realtime", "websockets"], "source": "/notebook/agentchat_realtime_webrtc.ipynb" }, { "title": "Agent with memory using Mem0", "link": "/docs/use-cases/notebooks/notebooks/agentchat_memory_using_mem0", "description": "Use Mem0 to create agents with memory.", "image": null, "tags": ["memory"], "source": "/notebook/agentchat_memory_using_mem0.ipynb" }, { "title": "Preprocessing Chat History with `TransformMessages`", "link": "/docs/use-cases/notebooks/notebooks/agentchat_transform_messages", "description": "Preprocessing chat history with `TransformMessages`", "image": null, "tags": ["long context handling", "capability"], "source": "/notebook/agentchat_transform_messages.ipynb" }, { "title": "RealtimeAgent in a Swarm Orchestration", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_swarm_websocket", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm"], "source": "/notebook/agentchat_realtime_swarm_websocket.ipynb" }, { "title": "RAG OpenAI Assistants in AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_oai_assistant_retrieval", "description": "OpenAI Assistant with retrieval augmentation.", "image": null, "tags": ["RAG", "OpenAI Assistant"], "source": "/notebook/agentchat_oai_assistant_retrieval.ipynb" }, { "title": "Using Guidance with AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_guidance", "description": "Constrained responses via guidance.", "image": null, "tags": ["guidance", "integration", "JSON"], "source": "/notebook/agentchat_guidance.ipynb" }, { "title": "Using RetrieveChat Powered by MongoDB Atlas for Retrieve Augmented Code Generation and Question Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_RetrieveChat_mongodb", "description": "Explore the use of AG2's RetrieveChat for tasks like code generation from docstrings, answering complex questions with human feedback, and exploiting features like Update Context, custom prompts, and few-shot learning.", "image": null, "tags": ["MongoDB", "integration", "RAG"], "source": "/notebook/agentchat_RetrieveChat_mongodb.ipynb" }, { "title": "Using Neo4j's graph database with AG2 agents for Question & Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_graph_rag_neo4j", "description": "Neo4j GraphRAG utilises a knowledge graph and can be added as a capability to agents.", "image": null, "tags": ["RAG"], "source": "/notebook/agentchat_graph_rag_neo4j.ipynb" }, { "title": "Use AG2 in Databricks with DBRX", "link": "/docs/use-cases/notebooks/notebooks/agentchat_databricks_dbrx", "description": "Use Databricks DBRX and Foundation Model APIs to build AG2 applications backed by open-source LLMs.", "image": null, "tags": ["integration", "code generation", "dbrx", "databricks", "open source", "lakehouse", "custom model", "data intelligence"], "source": "/notebook/agentchat_databricks_dbrx.ipynb" }, { "title": "Wikipedia Search Tools", "link": "/docs/use-cases/notebooks/notebooks/tools_wikipedia_search", "description": "Perplexity Search Integration with AG2", "image": null, "tags": ["tools", "perplexity", "web-search", "search"], "source": "/notebook/tools_wikipedia_search.ipynb" }, { "title": "Solving Complex Tasks with Nested Chats", "link": "/docs/use-cases/notebooks/notebooks/agentchat_nestedchat", "description": "Solve complex tasks with a chat nested as inner monologue.", "image": null, "tags": ["nested chat", "reflection", "reasoning", "orchestration"], "source": "/notebook/agentchat_nestedchat.ipynb" }, { "title": "DeepSeek: Adding Browsing Capabilities to AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_browser_use_deepseek", "description": "DeepSeek: Adding Browsing Capabilities to AG2", "image": null, "tags": ["tools", "browser-use", "webscraping", "function calling", "deepseek"], "source": "/notebook/tools_browser_use_deepseek.ipynb" }, { "title": "Solving Complex Tasks with A Sequence of Nested Chats", "link": "/docs/use-cases/notebooks/notebooks/agentchat_nested_sequential_chats", "description": "Solve complex tasks with one or more sequence chats nested as inner monologue.", "image": null, "tags": ["nested chat", "sequential chats", "orchestration"], "source": "/notebook/agentchat_nested_sequential_chats.ipynb" }, { "title": "Nested Chats for Tool Use in Conversational Chess", "link": "/docs/use-cases/notebooks/notebooks/agentchat_nested_chats_chess", "description": "LLM-backed agents playing chess with each other using nested chats.", "image": null, "tags": ["nested chat", "tool/function", "orchestration"], "source": "/notebook/agentchat_nested_chats_chess.ipynb" }, { "title": "OpenAI Assistants in AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_oai_assistant_twoagents_basic", "description": "Two-agent chat with OpenAI assistants.", "image": null, "tags": ["OpenAI Assistant"], "source": "/notebook/agentchat_oai_assistant_twoagents_basic.ipynb" }, { "title": "Group Chat with Coder and Visualization Critic", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_vis", "description": "Explore a group chat example using agents such as a coder and visualization agent.", "image": null, "tags": ["group chat", "code generation"], "source": "/notebook/agentchat_groupchat_vis.ipynb" }, { "title": "Agent Chat with Multimodal Models: LLaVA", "link": "/docs/use-cases/notebooks/notebooks/agentchat_lmm_llava", "description": "Leveraging multimodal models like llava.", "image": null, "tags": ["multimodal", "llava"], "source": "/notebook/agentchat_lmm_llava.ipynb" }, { "title": "SocietyOfMindAgent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_society_of_mind", "description": "Explore the demonstration of the SocietyOfMindAgent in the AG2 library, which runs a group chat as an internal monologue, but appears to the external world as a single agent, offering a structured way to manage complex interactions among multiple agents and handle issues such as extracting responses from complex dialogues and dealing with context window constraints.", "image": null, "tags": ["orchestration", "nested chat", "group chat"], "source": "/notebook/agentchat_society_of_mind.ipynb" }, { "title": "Conversational Workflows with MCP: A Marie Antoinette Take on The Eiffel Tower", "link": "/docs/use-cases/notebooks/notebooks/agentchat_mcp_filesystem", "description": "Conversational Workflows with MCP: A Marie Antoinette Take on The Eiffel Tower", "image": null, "tags": ["MCP"], "source": "/notebook/agentchat_mcp_filesystem.ipynb" }, { "title": "Agent Chat with Multimodal Models: DALLE and GPT-4V", "link": "/docs/use-cases/notebooks/notebooks/agentchat_dalle_and_gpt4v", "description": "Multimodal agent chat with DALL-E and GPT-4v.", "image": null, "tags": ["multimodal", "gpt-4v"], "source": "/notebook/agentchat_dalle_and_gpt4v.ipynb" }, { "title": "StateFlow: Build Workflows through State-Oriented Actions", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_stateflow", "description": "StateFlow: Build workflows through state-oriented actions.", "image": null, "tags": ["orchestration", "group chat", "stateflow", "research"], "source": "/notebook/agentchat_groupchat_stateflow.ipynb" }, { "title": "Small, Local Model (IBM Granite) Multi-Agent RAG", "link": "/docs/use-cases/notebooks/notebooks/agentchat_small_llm_rag_planning", "description": "Optimizing Small, Local LLMs in Multi-Agent RAG Workflows using IBM Granite, Document Retrieval, Web Search, and Ollama", "image": null, "tags": ["Small LLMs", "RAG", "Web Search", "IBM Granite", "Ollama", "Planning", "Reflection"], "source": "/notebook/agentchat_small_llm_rag_planning.ipynb" }, { "title": "Group Chat with Tools", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_tools", "description": "Group Chat with Tools", "image": null, "tags": ["agents", "tools", "group", "chat", "users", "guides"], "source": "/notebook/agentchat_groupchat_tools.ipynb" }, { "title": "DuckDuckGo Search Tool", "link": "/docs/use-cases/notebooks/notebooks/tools_duckduckgo_search", "description": "DuckDuckGo Search Tool", "image": null, "tags": ["tools", "DuckDuckGo", "web-search", "search"], "source": "/notebook/tools_duckduckgo_search.ipynb" }, { "title": "Load the configuration including the response format", "link": "/docs/use-cases/notebooks/notebooks/agentchat_quickstart_examples", "description": "Agent Quickstart Examples", "image": null, "tags": ["agents", "tools", "quickstart", "examples", "autogen"], "source": "/notebook/agentchat_quickstart_examples.ipynb" }, { "title": "RealtimeAgent in a Swarm Orchestration", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_gemini_swarm_websocket", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm"], "source": "/notebook/agentchat_realtime_gemini_swarm_websocket.ipynb" }, { "title": "MCP Clients", "link": "/docs/use-cases/notebooks/notebooks/mcp_client", "description": "MCP Clients", "image": null, "tags": ["MCP", "Model Context Protocol", "tools"], "source": "/notebook/mcp_client.ipynb" }, { "title": "Agent Tracking with AgentOps", "link": "/docs/use-cases/notebooks/notebooks/agentchat_agentops", "description": "Use AgentOps to simplify the development process and monitor your agents in production.", "image": null, "tags": ["integration", "monitoring", "debugging"], "source": "/notebook/agentchat_agentops.ipynb" }, { "title": "SQL Agent for Spider text-to-SQL benchmark", "link": "/docs/use-cases/notebooks/notebooks/agentchat_sql_spider", "description": "Natural language text to SQL query using the Spider text-to-SQL benchmark.", "image": null, "tags": ["SQL", "tool/function"], "source": "/notebook/agentchat_sql_spider.ipynb" }, { "title": "AutoBuild", "link": "/docs/use-cases/notebooks/notebooks/autobuild_basic", "description": "Automatically build multi-agent system with AgentBuilder", "image": null, "tags": ["autobuild"], "source": "/notebook/autobuild_basic.ipynb" }, { "title": "Automatically Build Multi-agent System from Agent Library", "link": "/docs/use-cases/notebooks/notebooks/autobuild_agent_library", "description": "Automatically build multi-agent system from agent library", "image": null, "tags": ["autobuild"], "source": "/notebook/autobuild_agent_library.ipynb" }, { "title": "Generate Dalle Images With Conversable Agents", "link": "/docs/use-cases/notebooks/notebooks/agentchat_image_generation_capability", "description": "Generate images with conversable agents.", "image": null, "tags": ["capability", "multimodal"], "source": "/notebook/agentchat_image_generation_capability.ipynb" }, { "title": "Agent Observability with OpenLIT", "link": "/docs/use-cases/notebooks/notebooks/agentchat_openlit", "description": "Use OpenLIT to easily monitor AI agents in production with OpenTelemetry.", "image": null, "tags": ["integration", "monitoring", "observability", "debugging"], "source": "/notebook/agentchat_openlit.ipynb" }, { "title": "Using RetrieveChat Powered by Couchbase Capella for Retrieve Augmented Code Generation and Question Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_RetrieveChat_couchbase", "description": "Explore the use of AG2's RetrieveChat for tasks like code generation from docstrings, answering complex questions with human feedback, and exploiting features like Update Context, custom prompts, and few-shot learning.", "image": null, "tags": ["RAG"], "source": "/notebook/agentchat_RetrieveChat_couchbase.ipynb" }, { "title": "A Uniform interface to call different LLMs", "link": "/docs/use-cases/notebooks/notebooks/autogen_uniformed_api_calling", "description": "Uniform interface to call different LLM.", "image": null, "tags": ["integration", "custom model"], "source": "/notebook/autogen_uniformed_api_calling.ipynb" }, { "title": "OptiGuide with Nested Chats in AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_nestedchat_optiguide", "description": "This is a nested chat re-implementation of OptiGuide which is an LLM-based supply chain optimization framework.", "image": null, "tags": ["nested chat", "hierarchical chat", "code generation", "orchestration"], "source": "/notebook/agentchat_nestedchat_optiguide.ipynb" }, { "title": "Conversational Chess using non-OpenAI clients", "link": "/docs/use-cases/notebooks/notebooks/agentchat_nested_chats_chess_altmodels", "description": "LLM-backed agents playing chess with each other using nested chats.", "image": null, "tags": ["nested chat", "tool/function", "orchestration"], "source": "/notebook/agentchat_nested_chats_chess_altmodels.ipynb" }, { "title": "Chat Context Dependency Injection", "link": "/docs/use-cases/notebooks/notebooks/tools_chat_context_dependency_injection", "description": "Chat Context Dependency Injection", "image": null, "tags": ["tools", "dependency injection", "function calling"], "source": "/notebook/tools_chat_context_dependency_injection.ipynb" }, { "title": "Agent with memory using Mem0", "link": "/docs/use-cases/notebooks/notebooks/agentchat_with_memory", "description": "Use Mem0 to create agents with memory.", "image": null, "tags": ["mem0", "integration", "memory"], "source": "/notebook/agentchat_with_memory.ipynb" }, { "title": "`run` function examples with event processing", "link": "/docs/use-cases/notebooks/notebooks/run_and_event_processing", "description": "Using run and event processing", "image": null, "tags": ["run", "event-processing", "integrate", "frontend"], "source": "/notebook/run_and_event_processing.ipynb" }, { "title": "From Dad Jokes To Sad Jokes: Function Calling with GPTAssistantAgent", "link": "/docs/use-cases/notebooks/notebooks/gpt_assistant_agent_function_call", "description": "Use tools in a GPTAssistantAgent Multi-Agent System by utilizing functions such as calling an API and writing to a file.", "image": null, "tags": ["OpenAI Assistant", "tool/function"], "source": "/notebook/gpt_assistant_agent_function_call.ipynb" }, { "title": "Mitigating Prompt hacking with JSON Mode in Autogen", "link": "/docs/use-cases/notebooks/notebooks/JSON_mode_example", "description": "Use JSON mode and Agent Descriptions to mitigate prompt manipulation and control speaker transition.", "image": null, "tags": ["JSON", "description", "prompt hacking", "group chat", "orchestration"], "source": "/notebook/JSON_mode_example.ipynb" }, { "title": "Group Chat with Customized Speaker Selection Method", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_customized", "description": "Introduce group chat with customized speaker selection method.", "image": null, "tags": ["orchestration", "group chat"], "source": "/notebook/agentchat_groupchat_customized.ipynb" }, { "title": "Trip planning with a FalkorDB GraphRAG agent using a Swarm", "link": "/docs/use-cases/notebooks/notebooks/agentchat_swarm_graphrag_trip_planner", "description": "FalkorDB GraphRAG utilises a knowledge graph and can be added as a capability to agents. Together with a swarm orchestration of agents is highly effective at providing a RAG capability.", "image": null, "tags": ["RAG", "tool/function", "swarm"], "source": "/notebook/agentchat_swarm_graphrag_trip_planner.ipynb" }, { "title": "Tavily Search Tool", "link": "/docs/use-cases/notebooks/notebooks/tools_tavily_search", "description": "Tavily Search Integration with AG2", "image": null, "tags": ["tools", "tavily", "web-search", "search"], "source": "/notebook/tools_tavily_search.ipynb" }, { "title": "Structured output", "link": "/docs/use-cases/notebooks/notebooks/agentchat_structured_outputs", "description": "LLM providers offer functionality for defining a structure of the messages generated by LLMs, AG2 enables this functionality by propagating a `response_format`, in the LLM configuration for your agents, to the underlying client.", "image": null, "tags": ["structured output"], "source": "/notebook/agentchat_structured_outputs.ipynb" }, { "title": "Adding Google Search Capability to AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_google_search", "description": "Google Search", "image": null, "tags": ["agents", "tools", "search", "web", "google", "real-time search"], "source": "/notebook/tools_google_search.ipynb" }, { "title": "Task Solving with Provided Tools as Functions (Asynchronous Function Calls)", "link": "/docs/use-cases/notebooks/notebooks/agentchat_function_call_async", "description": "Learn how to implement both synchronous and asynchronous function calls using AssistantAgent and UserProxyAgent in AutoGen, with examples of their application in individual and group chat settings for task execution with language models.", "image": null, "tags": ["tool/function", "async"], "source": "/notebook/agentchat_function_call_async.ipynb" }, { "title": "Conversational Workflows with MCP: A French joke on a random Wikipedia article", "link": "/docs/use-cases/notebooks/notebooks/agentchat_mcp_wikipedia", "description": "Conversational Workflows with MCP: A French joke on a random Wikipedia article", "image": null, "tags": ["MCP"], "source": "/notebook/agentchat_mcp_wikipedia.ipynb" }, { "title": "Auto Generated Agent Chat: Using MathChat to Solve Math Problems", "link": "/docs/use-cases/notebooks/notebooks/agentchat_MathChat", "description": "Using MathChat to Solve Math Problems", "image": null, "tags": ["math"], "source": "/notebook/agentchat_MathChat.ipynb" }, { "title": "Agent Chat with custom model loading", "link": "/docs/use-cases/notebooks/notebooks/agentchat_custom_model", "description": "Define and load a custom model", "image": null, "tags": ["integration", "custom model"], "source": "/notebook/agentchat_custom_model.ipynb" }, { "title": "Auto Generated Agent Chat: Function Inception", "link": "/docs/use-cases/notebooks/notebooks/agentchat_inception_function", "description": "Function Inception: Enable AG2 agents to update/remove functions during conversations.", "image": null, "tags": ["function inception", "tool/function"], "source": "/notebook/agentchat_inception_function.ipynb" }, { "title": "Use AG2 to Tune ChatGPT", "link": "/docs/use-cases/notebooks/notebooks/oai_chatgpt_gpt4", "description": "Use AG2 to Tune ChatGPT", "image": null, "tags": ["llm", "hyperparamater", "tuning", "gpt", "parameter tuning"], "source": "/notebook/oai_chatgpt_gpt4.ipynb" }, { "title": "Auto Generated Agent Chat: Group Chat with GPTAssistantAgent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_oai_assistant_groupchat", "description": "Use GPTAssistantAgent in group chat.", "image": null, "tags": ["OpenAI Assistant", "group chat"], "source": "/notebook/agentchat_oai_assistant_groupchat.ipynb" }, { "title": "Using FalkorGraphRagCapability with agents for GraphRAG Question & Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_graph_rag_falkordb", "description": "Using FalkorGraphRagCapability with agents for GraphRAG Question & Answering", "image": null, "tags": ["RAG", "FalkorDB"], "source": "/notebook/agentchat_graph_rag_falkordb.ipynb" }, { "title": "Solving Multiple Tasks in a Sequence of Async Chats", "link": "/docs/use-cases/notebooks/notebooks/agentchat_multi_task_async_chats", "description": "Use conversational agents to solve a set of tasks with a sequence of async chats.", "image": null, "tags": ["orchestration", "async", "sequential chats"], "source": "/notebook/agentchat_multi_task_async_chats.ipynb" }, { "title": "Discord, Slack, and Telegram messaging tools", "link": "/docs/use-cases/notebooks/notebooks/tools_commsplatforms", "description": "Adding Browsing Capabilities to AG2", "image": null, "tags": ["tools", "browser-use", "webscraping", "function calling"], "source": "/notebook/tools_commsplatforms.ipynb" }, { "title": "RealtimeAgent in a Swarm Orchestration using WebRTC", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_swarm_webrtc", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm"], "source": "/notebook/agentchat_realtime_swarm_webrtc.ipynb" }, { "title": "Runtime Logging with AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_logging", "description": "Provide capabilities of runtime logging for debugging and performance analysis.", "image": null, "tags": ["logging", "debugging"], "source": "/notebook/agentchat_logging.ipynb" }, { "title": "WebSurferAgent", "link": "/docs/use-cases/notebooks/notebooks/agents_websurfer", "description": "WebSurfer Agent", "image": null, "tags": ["agents", "browser-use", "crawl4ai", "webscraping", "function calling"], "source": "/notebook/agents_websurfer.ipynb" }, { "title": "Use MongoDBQueryEngine to query Markdown files", "link": "/docs/use-cases/notebooks/notebooks/mongodb_query_engine", "description": "Mongo DB Query Engine", "image": null, "tags": ["agents", "documents", "RAG", "docagent", "mongodb", "query"], "source": "/notebook/mongodb_query_engine.ipynb" }, { "title": "Conversational Workflows with MCP: A Shakespearean Take on arXiv Abstracts", "link": "/docs/use-cases/notebooks/notebooks/agentchat_mcp_arxiv", "description": "Conversational Workflows with MCP: A Shakespearean Take on arXiv Abstracts", "image": null, "tags": ["MCP"], "source": "/notebook/agentchat_mcp_arxiv.ipynb" }, { "title": "RAG with DocAgent", "link": "/docs/use-cases/notebooks/notebooks/agents_docagent", "description": "Query documents and web pages with DocAgent", "image": null, "tags": ["agents", "documents", "RAG", "docagent"], "source": "/notebook/agents_docagent.ipynb" }, { "title": "Solving Multiple Tasks in a Sequence of Chats", "link": "/docs/use-cases/notebooks/notebooks/agentchat_multi_task_chats", "description": "Use conversational agents to solve a set of tasks with a sequence of chats.", "image": null, "tags": ["orchestration", "sequential chats"], "source": "/notebook/agentchat_multi_task_chats.ipynb" }, { "title": "Cross-Framework LLM Tool for CaptainAgent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_captainagent_crosstool", "description": "Cross-Framework LLM Tool for CaptainAgent", "image": null, "tags": ["tools", "langchain", "crewai"], "source": "/notebook/agentchat_captainagent_crosstool.ipynb" }, { "title": "Web Scraping using Apify Tools", "link": "/docs/use-cases/notebooks/notebooks/agentchat_webscraping_with_apify", "description": "Scrapping web pages and summarizing the content using agents with tools.", "image": null, "tags": ["web", "apify", "integration", "tool/function"], "source": "/notebook/agentchat_webscraping_with_apify.ipynb" }, { "title": "Auto Generated Agent Chat: Collaborative Task Solving with Coding and Planning Agent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_planning", "description": "Use planning agent in a function call.", "image": null, "tags": ["planning", "orchestration", "nested chat", "tool/function"], "source": "/notebook/agentchat_planning.ipynb" }, { "title": "Currency Calculator: Task Solving with Provided Tools as Functions", "link": "/docs/use-cases/notebooks/notebooks/agentchat_function_call_currency_calculator", "description": "Learn how to register function calls using AssistantAgent and UserProxyAgent.", "image": null, "tags": ["tool/function"], "source": "/notebook/agentchat_function_call_currency_calculator.ipynb" }, { "title": "Use ChromaDBQueryEngine to query Markdown files", "link": "/docs/use-cases/notebooks/notebooks/Chromadb_query_engine", "description": "ChromaDB Query Engine for document queries", "image": null, "tags": ["agents", "documents", "RAG", "docagent", "chroma", "chromadb"], "source": "/notebook/Chromadb_query_engine.ipynb" }, { "title": "Using RetrieveChat for Retrieve Augmented Code Generation and Question Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_RetrieveChat", "description": "Explore the use of AG2's RetrieveChat for tasks like code generation from docstrings, answering complex questions with human feedback, and exploiting features like Update Context, custom prompts, and few-shot learning.", "image": null, "tags": ["RAG"], "source": "/notebook/agentchat_RetrieveChat.ipynb" }, { "title": "Writing a software application using function calls", "link": "/docs/use-cases/notebooks/notebooks/agentchat_function_call_code_writing", "description": "Equip your agent with functions that can efficiently implement features into your software application.", "image": null, "tags": ["code generation", "tool/function", "fastapi", "software engineering"], "source": "/notebook/agentchat_function_call_code_writing.ipynb" }, { "title": "Using OpenAI\u2019s Web Search Tool with AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_web_search_preview", "description": "Web Search Preview", "image": null, "tags": ["tools", "search", "responses api"], "source": "/notebook/tools_web_search_preview.ipynb" }, { "title": "Using RetrieveChat Powered by PGVector for Retrieve Augmented Code Generation and Question Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_RetrieveChat_pgvector", "description": "Explore the use of AG2's RetrieveChat for tasks like code generation from docstrings, answering complex questions with human feedback, and exploiting features like Update Context, custom prompts, and few-shot learning.", "image": null, "tags": ["PGVector", "integration", "RAG"], "source": "/notebook/agentchat_RetrieveChat_pgvector.ipynb" }, { "title": "Enhanced Swarm Orchestration with AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_swarm_enhanced", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm"], "source": "/notebook/agentchat_swarm_enhanced.ipynb" }, { "title": "Perform Research with Multi-Agent Group Chat", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat_research", "description": "Perform research using a group chat with a number of specialized agents.", "image": null, "tags": ["group chat", "planning", "code generation"], "source": "/notebook/agentchat_groupchat_research.ipynb" }, { "title": "Auto Generated Agent Chat: Teaching AI New Skills via Natural Language Interaction", "link": "/docs/use-cases/notebooks/notebooks/agentchat_teaching", "description": "Teach the agent news skills using natural language.", "image": null, "tags": ["learning", "teaching"], "source": "/notebook/agentchat_teaching.ipynb" }, { "title": "Usage tracking with AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_cost_token_tracking", "description": "cost calculation", "image": null, "tags": ["cost"], "source": "/notebook/agentchat_cost_token_tracking.ipynb" }, { "title": "Using Neo4j's native GraphRAG SDK with AG2 agents for Question & Answering", "link": "/docs/use-cases/notebooks/notebooks/agentchat_graph_rag_neo4j_native", "description": "Neo4j Native GraphRAG utilizes a knowledge graph and can be added as a capability to agents.", "image": null, "tags": ["RAG"], "source": "/notebook/agentchat_graph_rag_neo4j_native.ipynb" }, { "title": "Groupchat with Llamaindex agents", "link": "/docs/use-cases/notebooks/notebooks/agentchat_group_chat_with_llamaindex_agents", "description": "Integrate llamaindex agents with Autogen.", "image": null, "tags": ["react", "llamaindex", "integration", "group chat", "software engineering"], "source": "/notebook/agentchat_group_chat_with_llamaindex_agents.ipynb" }, { "title": "Assistants with Azure Cognitive Search and Azure Identity", "link": "/docs/use-cases/notebooks/notebooks/agentchat_azr_ai_search", "description": "This notebook demonstrates the use of Assistant Agents in conjunction with Azure Cognitive Search and Azure Identity", "image": null, "tags": ["integration", "RAG", "Azure Identity", "Azure AI Search"], "source": "/notebook/agentchat_azr_ai_search.ipynb" }, { "title": "Swarm Orchestration with AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_swarm", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm"], "source": "/notebook/agentchat_swarm.ipynb" }, { "title": "Tools with Dependency Injection", "link": "/docs/use-cases/notebooks/notebooks/tools_dependency_injection", "description": "Tools Dependency Injection", "image": null, "tags": ["tools", "dependency injection", "function calling"], "source": "/notebook/tools_dependency_injection.ipynb" }, { "title": "Structured output from json configuration", "link": "/docs/use-cases/notebooks/notebooks/agentchat_structured_outputs_from_config", "description": "OpenAI offers a functionality for defining a structure of the messages generated by LLMs, AutoGen enables this functionality by propagating response_format passed to your agents to the underlying client.", "image": null, "tags": ["structured output"], "source": "/notebook/agentchat_structured_outputs_from_config.ipynb" }, { "title": "Solving Multiple Tasks in a Sequence of Chats with Different Conversable Agent Pairs", "link": "/docs/use-cases/notebooks/notebooks/agentchats_sequential_chats", "description": "Use AG2 to solve a set of tasks with a sequence of chats.", "image": null, "tags": ["orchestration", "sequential chats"], "source": "/notebook/agentchats_sequential_chats.ipynb" }, { "title": "Chat with OpenAI Assistant using function call in AG2: OSS Insights for Advanced GitHub Data Analysis", "link": "/docs/use-cases/notebooks/notebooks/agentchat_oai_assistant_function_call", "description": "This Jupyter Notebook demonstrates how to leverage OSS Insight (Open Source Software Insight) for advanced GitHub data analysis by defining `Function calls` in AG2 for the OpenAI Assistant.", "image": null, "tags": ["OpenAI Assistant", "tool/function"], "source": "/notebook/agentchat_oai_assistant_function_call.ipynb" }, { "title": "Auto Generated Agent Chat: Collaborative Task Solving with Multiple Agents and Human Users", "link": "/docs/use-cases/notebooks/notebooks/agentchat_two_users", "description": "Involve multiple human users via function calls and nested chat.", "image": null, "tags": ["human", "tool/function"], "source": "/notebook/agentchat_two_users.ipynb" }, { "title": "Adding YouTube Search Capability to AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_youtube_search", "description": "YouTube Search Integration with AG2", "image": null, "tags": ["tools", "youtube", "video", "search"], "source": "/notebook/tools_youtube_search.ipynb" }, { "title": "Chatting with a teachable agent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_teachability", "description": "Learn how to persist memories across chat sessions using the Teachability capability", "image": null, "tags": ["teachability", "learning", "RAG", "capability"], "source": "/notebook/agentchat_teachability.ipynb" }, { "title": "Making OpenAI Assistants Teachable", "link": "/docs/use-cases/notebooks/notebooks/agentchat_teachable_oai_assistants", "description": "Teach OpenAI assistants.", "image": null, "tags": ["teachability", "capability", "learning", "RAG", "OpenAI Assistant"], "source": "/notebook/agentchat_teachable_oai_assistants.ipynb" }, { "title": "RealtimeAgent in a Swarm Orchestration", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_swarm", "description": "Swarm Ochestration", "image": null, "tags": ["orchestration", "group chat", "swarm", "realtime"], "source": "/notebook/agentchat_realtime_swarm.ipynb" }, { "title": "Translating Video audio using Whisper and GPT-3.5-turbo", "link": "/docs/use-cases/notebooks/notebooks/agentchat_video_transcript_translate_with_whisper", "description": "Use tools to extract and translate the transcript of a video file.", "image": null, "tags": ["whisper", "multimodal", "tool/function"], "source": "/notebook/agentchat_video_transcript_translate_with_whisper.ipynb" }, { "title": "Run a standalone AssistantAgent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_assistant_agent_standalone", "description": "Run a standalone AssistantAgent, browsing the web using the BrowserUseTool", "image": null, "tags": ["assistantagent", "run", "browser-use", "webscraping", "function calling"], "source": "/notebook/agentchat_assistant_agent_standalone.ipynb" }, { "title": "Auto Generated Agent Chat: Task Solving with Langchain Provided Tools as Functions", "link": "/docs/use-cases/notebooks/notebooks/agentchat_langchain", "description": "Use Langchain tools.", "image": null, "tags": ["langchain", "integration", "tool/function"], "source": "/notebook/agentchat_langchain.ipynb" }, { "title": "Use LLamaIndexQueryEngine to query Markdown files", "link": "/docs/use-cases/notebooks/notebooks/LlamaIndex_query_engine", "description": "Use any LlamaIndex vector store as a Query Engine", "image": null, "tags": ["agents", "documents", "RAG", "docagent", "chroma", "chromadb", "pinecone"], "source": "/notebook/LlamaIndex_query_engine.ipynb" }, { "title": "Auto Generated Agent Chat: GPTAssistant with Code Interpreter", "link": "/docs/use-cases/notebooks/notebooks/agentchat_oai_code_interpreter", "description": "This Jupyter Notebook showcases the integration of the Code Interpreter tool which executes Python code dynamically within applications.", "image": null, "tags": ["OpenAI Assistant", "code interpreter"], "source": "/notebook/agentchat_oai_code_interpreter.ipynb" }, { "title": "Interactive LLM Agent Dealing with Data Stream", "link": "/docs/use-cases/notebooks/notebooks/agentchat_stream", "description": "Automated continual learning from new data.", "image": null, "tags": ["streaming", "async", "learning"], "source": "/notebook/agentchat_stream.ipynb" }, { "title": "Agent Chat with Async Human Inputs", "link": "/docs/use-cases/notebooks/notebooks/async_human_input", "description": "Async human inputs.", "image": null, "tags": ["async", "human"], "source": "/notebook/async_human_input.ipynb" }, { "title": "ReasoningAgent - Advanced LLM Reasoning with Multiple Search Strategies", "link": "/docs/use-cases/notebooks/notebooks/agentchat_reasoning_agent", "description": "Use ReasoningAgent for o1 style reasoning in Agentic workflows with LLMs using AG2", "image": null, "tags": ["reasoning agent", "tree of thoughts"], "source": "/notebook/agentchat_reasoning_agent.ipynb" }, { "title": "Auto Generated Agent Chat: Solving Tasks Requiring Web Info", "link": "/docs/use-cases/notebooks/notebooks/agentchat_web_info", "description": "Solve tasks requiring web info.", "image": null, "tags": ["web", "code generation"], "source": "/notebook/agentchat_web_info.ipynb" }, { "title": "Use AG2 to Tune OpenAI Models", "link": "/docs/use-cases/notebooks/notebooks/oai_completion", "description": "Use AG2 to Tune OpenAI Models", "image": null, "tags": ["llm", "hyperparamater", "tuning", "openai", "gpt", "parameter tuning"], "source": "/notebook/oai_completion.ipynb" }, { "title": "Engaging with Multimodal Models: GPT-4V in AG2", "link": "/docs/use-cases/notebooks/notebooks/agentchat_lmm_gpt-4v", "description": "Leveraging multimodal models through two different methodologies: MultimodalConversableAgent and VisionCapability.", "image": null, "tags": ["multimodal", "gpt-4v"], "source": "/notebook/agentchat_lmm_gpt-4v.ipynb" }, { "title": "Supercharging Web Crawling with Crawl4AI", "link": "/docs/use-cases/notebooks/notebooks/tools_crawl4ai", "description": "Supercharging Web Crawling with Crawl4AI", "image": null, "tags": ["tools", "browser-use", "webscraping", "function calling"], "source": "/notebook/tools_crawl4ai.ipynb" }, { "title": "Use AG2 in Microsoft Fabric", "link": "/docs/use-cases/notebooks/notebooks/agentchat_microsoft_fabric", "description": "Use AG2 in Microsoft Fabric", "image": null, "tags": ["agents", "chat", "microsoft", "fabric", "guides"], "source": "/notebook/agentchat_microsoft_fabric.ipynb" }, { "title": "Cross-Framework LLM Tool Integration with AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_interoperability", "description": "Cross-Framework LLM Tool Integration with AG2", "image": null, "tags": ["tools", "langchain", "crewai", "pydanticai"], "source": "/notebook/tools_interoperability.ipynb" }, { "title": "Demonstrating the `AgentEval` framework using the task of solving math problems as an example", "link": "/docs/use-cases/notebooks/notebooks/agenteval_cq_math", "description": "AgentEval: a multi-agent system for assessing utility of LLM-powered applications", "image": null, "tags": ["eval"], "source": "/notebook/agenteval_cq_math.ipynb" }, { "title": "Group Chat", "link": "/docs/use-cases/notebooks/notebooks/agentchat_groupchat", "description": "Explore the utilization of large language models in automated group chat scenarios, where agents perform tasks collectively, demonstrating how they can be configured, interact with each other, and retrieve specific information from external resources.", "image": null, "tags": ["orchestration", "group chat", "code generation"], "source": "/notebook/agentchat_groupchat.ipynb" }, { "title": "Adding Browsing Capabilities to AG2", "link": "/docs/use-cases/notebooks/notebooks/tools_browser_use", "description": "Adding Browsing Capabilities to AG2", "image": null, "tags": ["tools", "browser-use", "webscraping", "function calling"], "source": "/notebook/tools_browser_use.ipynb" }, { "title": "CaptainAgent", "link": "/docs/use-cases/notebooks/notebooks/agentchat_captainagent", "description": "Introducing CaptainAgent, a powerful agent that can manage and orchestrate other agents and tools to solve complex tasks.", "image": null, "tags": ["autobuild", "CaptainAgent"], "source": "/notebook/agentchat_captainagent.ipynb" }, { "title": "(Legacy) Implement Swarm-style orchestration with GroupChat", "link": "/docs/use-cases/notebooks/notebooks/agentchat_swarm_w_groupchat_legacy", "description": "(Legacy) Implement Swarm-style orchestration with GroupChat", "image": null, "tags": ["orchestration", "group chat", "stateflow", "swarm"], "source": "/notebook/agentchat_swarm_w_groupchat_legacy.ipynb" }, { "title": "Task Solving with Code Generation, Execution and Debugging", "link": "/docs/use-cases/notebooks/notebooks/agentchat_auto_feedback_from_code_execution", "description": "Use conversable language learning model agents to solve tasks and provide automatic feedback through a comprehensive example of writing, executing, and debugging Python code to compare stock price changes.", "image": null, "tags": ["code generation"], "source": "/notebook/agentchat_auto_feedback_from_code_execution.ipynb" }, { "title": "RealtimeAgent with local websocket connection", "link": "/docs/use-cases/notebooks/notebooks/agentchat_realtime_websocket", "description": "RealtimeAgent using websockets", "image": null, "tags": ["realtime", "websockets"], "source": "/notebook/agentchat_realtime_websocket.ipynb" }]; export const ClientSideComponent = ({Component, componentProps}) => { if (typeof document === "undefined") { return null; } return ; }; export const GalleryPage = ({galleryItems, allowDefaultImage = true, target = "_blank"}) => { const getTagsFromURL = () => { const searchParams = new URLSearchParams(window.location.search); const tags = searchParams.get("tags"); return tags ? tags.split(",") : []; }; let selectedTags = getTagsFromURL(); const defaultImageIfNoImage = allowDefaultImage ?? true; const allTags = [...new Set(galleryItems.flatMap(item => item.tags))]; const handleCardClick = (e, targetLink) => { if (!e.target.closest('a')) { if (target === '_blank') { window.open(targetLink, '_blank'); } else { window.location.href = targetLink; } } }; const updateURL = tags => { const searchParams = new URLSearchParams(window.location.search); if (tags.length > 0) { searchParams.set("tags", tags.join(",")); } else { searchParams.delete("tags"); } const searchParamsString = searchParams.toString(); const newURL = searchParamsString ? `${window.location.pathname}?${searchParamsString}` : window.location.pathname; window.history.pushState({}, '', newURL); }; window.addEventListener('popstate', () => { selectedTags = getTagsFromURL(); filterItems(); const select = document.querySelector('.tag-filter'); if (select && window.jQuery) { $(select).val(selectedTags).trigger('chosen:updated'); } }); const handleGalleryTagChange = event => { handleTagChange(event.detail); }; document.removeEventListener('gallery:tagChange', handleGalleryTagChange); document.addEventListener('gallery:tagChange', handleGalleryTagChange); const imageFunc = item => { const image = {item.title}; const imageToUse = item.image ? image : defaultImageIfNoImage ? image : null; return imageToUse; }; const handleTagChange = tags => { selectedTags = tags; updateURL(tags); filterItems(); }; const filterItems = () => { const cards = document.querySelectorAll('.examples-gallery-container .card'); cards.forEach(card => { const cardTags = Array.from(card.querySelectorAll('.tag')).map(tag => tag.textContent); if (selectedTags.length === 0 || selectedTags.some(tag => cardTags.includes(tag))) { card.style.display = ''; } else { card.style.display = 'none'; } }); }; const TagsView = ({tags}) =>
{tags?.map((tag, index) => { if (!selectedTags.includes(tag)) { handleTagChange([...selectedTags, tag]); const select = document.querySelector('.tag-filter'); if (select && window.jQuery) { $(select).val([...selectedTags, tag]).trigger('chosen:updated'); } } evt.preventDefault(); evt.stopPropagation(); return false; }}> {tag} )}
; const badges = item => { if (!item.source) { return null; } const colab_href = `https://colab.research.google.com/github/ag2ai/ag2/blob/main/${item.source}`; const github_href = `https://github.com/ag2ai/ag2/blob/main/${item.source}`; return Open In Colab Static Badge ; }; return
{galleryItems.map((item, index) =>
handleCardClick(e, item.link)}> {imageFunc(item)}
{item.title}
{badges(item)}

{item.description || item.title}

)}
; }; This page contains a collection of notebooks that demonstrate how to use AG2. The notebooks are tagged with the topics they cover. For example, a notebook that demonstrates how to use function calling will be tagged with `tool/function`.
Edit this page