Glossary

Relevant source files

• README.md
• README_ar.md
• README_fr.md
• README_id.md
• README_ja.md
• README_ko.md
• README_pt_br.md
• README_tr.md
• README_tzh.md
• README_zh.md
• admin/server/admin_server.py
• agent/canvas.py
• agent/component/agent_with_tools.py
• agent/component/base.py
• agent/component/categorize.py
• agent/component/llm.py
• agent/component/message.py
• agent/tools/base.py
• api/apps/__init__.py
• api/apps/backward_compat.py
• api/apps/llm_app.py
• api/apps/restful_apis/chunk_api.py
• api/apps/restful_apis/dataset_api.py
• api/apps/services/dataset_api_service.py
• api/db/__init__.py
• api/db/db_models.py
• api/db/init_data.py
• api/db/services/dialog_service.py
• api/db/services/document_service.py
• api/db/services/knowledgebase_service.py
• api/db/services/llm_service.py
• api/db/services/task_service.py
• api/db/services/user_service.py
• api/ragflow_server.py
• api/settings.py
• api/utils/api_utils.py
• api/utils/validation_utils.py
• common/doc_store/infinity_conn_base.py
• common/mcp_tool_call_conn.py
• conf/infinity_mapping.json
• conf/llm_factories.json
• deepdoc/parser/pdf_parser.py
• deepdoc/vision/layout_recognizer.py
• deepdoc/vision/ocr.py
• deepdoc/vision/recognizer.py
• deepdoc/vision/t_ocr.py
• deepdoc/vision/t_recognizer.py
• deepdoc/vision/table_structure_recognizer.py
• docs/guides/manage_files.md
• docs/quickstart.mdx
• docs/references/http_api_reference.md
• docs/references/python_api_reference.md
• docs/release_notes.md
• internal/admin/router.go
• internal/cli/admin_command.go
• internal/cli/admin_parser.go
• internal/cli/cli.go
• internal/cli/cli_http.go
• internal/cli/common_command.go
• internal/cli/lexer.go
• internal/cli/parser.go
• internal/cli/types.go
• internal/cli/user_command.go
• internal/cli/user_parser.go
• internal/dao/document.go
• internal/dao/document_test.go
• internal/handler/document.go
• internal/handler/document_test.go
• internal/router/router.go
• internal/service/nlp/retrieval.go
• rag/advanced_rag/knowlege_compile/_common.py
• rag/advanced_rag/knowlege_compile/dataset_nav.py
• rag/advanced_rag/knowlege_compile/raptor.py
• rag/advanced_rag/knowlege_compile/runner.py
• rag/advanced_rag/knowlege_compile/structure.py
• rag/advanced_rag/knowlege_compile/wiki.py
• rag/flow/compiler/compiler.py
• rag/llm/__init__.py
• rag/llm/chat_model.py
• rag/llm/cv_model.py
• rag/llm/embedding_model.py
• rag/llm/rerank_model.py
• rag/llm/sequence2txt_model.py
• rag/llm/tts_model.py
• rag/nlp/query.py
• rag/nlp/rag_tokenizer.py
• rag/nlp/search.py
• rag/nlp/term_weight.py
• rag/prompts/generator.py
• rag/settings.py
• rag/svr/task_executor.py
• rag/svr/task_executor_refactor/chunk_post_processor.py
• rag/svr/task_executor_refactor/dataflow_service.py
• rag/svr/task_executor_refactor/dataset_structure_merger.py
• rag/svr/task_executor_refactor/dataset_wiki_generator.py
• rag/svr/task_executor_refactor/task_handler.py
• rag/utils/es_conn.py
• rag/utils/infinity_conn.py
• rag/utils/redis_conn.py
• sdk/python/ragflow_sdk/modules/__init__.py
• sdk/python/ragflow_sdk/modules/agent.py
• sdk/python/ragflow_sdk/modules/base.py
• sdk/python/ragflow_sdk/modules/chat.py
• sdk/python/ragflow_sdk/modules/chunk.py
• sdk/python/ragflow_sdk/modules/dataset.py
• sdk/python/ragflow_sdk/modules/document.py
• sdk/python/ragflow_sdk/modules/session.py
• sdk/python/ragflow_sdk/ragflow.py
• test/testcases/test_web_api/test_chunk_app/test_chunk_routes_unit.py
• test/testcases/test_web_api/test_system_app/test_apps_init_unit.py
• test/unit_test/rag/test_raptor_psi_tree_builder.py
• web/src/components/document-preview/md/index.tsx
• web/src/components/svg-icon.tsx
• web/src/components/ui/tree-view.tsx
• web/src/constants/llm.ts
• web/src/constants/markdown-remark-plugins.ts
• web/src/utils/common-util.ts

This page provides definitions and technical pointers for codebase-specific terms, domain concepts, and architectural jargon used within RAGFlow.

Purpose and Scope

The RAGFlow glossary serves as a reference for onboarding engineers to understand the specific nomenclature used in the Python backend, the Go server, and the React frontend. It bridges high-level RAG (Retrieval-Augmented Generation) concepts with their specific implementations in the rag/, api/, and agent/ directories.

Core Domain Terms

RAG (Retrieval-Augmented Generation)

The core architectural pattern of the system. It involves retrieving relevant document chunks from a data store to provide context for a Large Language Model (LLM) to generate grounded responses README.md78

Dataset (Knowledge Base)

A logical collection of documents belonging to a tenant. In the code, this is often referred to as Knowledgebase or KB.

• Code Entity: KnowledgebaseService api/db/services/knowledgebase_service.py35
• Database Table: Knowledgebase api/db/db_models.py26
• API Layer: Managed via datasetsHandler in the Go router internal/router/router.go31

Chunking (Parsing)

The process of breaking down a document into smaller, searchable units. RAGFlow uses various parsers based on document structure.

• Code Entity: ParserType common/constants.py76
• Implementation: The FACTORY in task_executor.py maps parser types (e.g., naive, paper, book, qa, laws) to specific implementation modules rag/svr/task_executor.py114-131

Embedding

The process of converting text chunks into numerical vectors. These vectors are stored in a document store for similarity search.

• Code Entity: Base embedding class rag/llm/embedding_model.py145
• Token Ceiling: Standard token ceiling for common 8K-context embedding models is defined by DEFAULT_MAX_TOKENS rag/llm/embedding_model.py44

Reranking

A second-stage retrieval process where a specialized model (Reranker) scores the relevance of retrieved chunks against the query to improve precision.

• Code Entity: Base rerank class rag/llm/rerank_model.py28
• Integration: Used in the search pipeline to improve recall accuracy rag/nlp/search.py134

Codebase-Specific Jargon

Dialog (Assistant)

A configuration object that defines how a chat behaves, including which datasets it searches, which LLM it uses, and its system prompt.

• Code Entity: DialogService api/db/services/dialog_service.py140-141
• Database Table: Dialog api/db/db_models.py32

Session (Conversation)

A specific instance of an interaction between a user and a Dialog or Agent. It maintains the message history.

• Code Entity: ConversationService api/db/services/conversation_service.py26 (referenced in api/db/services/dialog_service.py178-183)

Tenant

The top-level entity for multi-tenancy. All datasets, documents, and LLM configurations are scoped to a tenant_id.

• Model Wrapper: LLMBundle api/db/services/llm_service.py79
• Default Models: get_tenant_default_model_by_type retrieves tenant-specific LLM settings api/db/joint_services/tenant_model_service.py82

Task Executor

The background worker process that handles heavy lifting like document parsing, embedding, and GraphRAG indexing. It consumes tasks from Redis Streams.

• Key Function: set_progress() updates task status and handles cancellation rag/svr/task_executor.py189-210
• Task Types: Maps identifiers like dataflow, raptor, memory to PipelineTaskType rag/svr/task_executor.py133-147

Memory

A feature for AI agents to persist information across sessions, categorized into types like raw, semantic, episodic, and procedural.

• Code Entity: handle_save_to_memory_task api/db/joint_services/memory_message_service.py44
• Task Type: PipelineTaskType.MEMORY rag/svr/task_executor.py138

System Space Mapping

Natural Language to Code Entity Mapping

This diagram shows how user-facing concepts in the UI/Docs map to specific classes and database models in the backend.

Concept Mapping Diagram

Sources: api/db/db_models.py25-32 api/db/services/knowledgebase_service.py35 api/db/services/dialog_service.py140-141 api/db/services/document_service.py42-43

Task Execution Data Flow

Mapping the lifecycle of a document upload from the "Natural Language Space" (User Action) to the "Code Entity Space" (Background Processing).

Task Lifecycle Diagram

Sources: rag/svr/task_executor.py114-131 rag/svr/task_executor.py189-210 api/db/services/document_service.py42-43 api/db/services/task_service.py80-81 common/constants.py103 docs/references/http_api_reference.md42

Technical Abbreviations

Abbreviation	Full Name	Description	Code Pointer
DSL	Domain Specific Language	The JSON representation of an agent workflow graph.	agent/canvas.py29
MCP	Model Context Protocol	Protocol for connecting LLMs to external data and tools.	internal/router/router.go45-46
RAPTOR	Recursive Abstractive Processing...	A tree-organized retrieval method for long documents.	rag/svr/task_executor.py48-56
OCR	Optical Character Recognition	Used in document parsing to extract text from images or PDFs.	rag/svr/task_executor.py125
KG	Knowledge Graph	Graph-based parsing and retrieval strategy.	rag/svr/task_executor.py129
ReAct	Reasoning and Acting	A prompting strategy for agents to use tools iteratively.	rag/llm/chat_model.py59-61
TTS	Text To Speech	Model type for generating audio from text.	rag/llm/chat_model.py37
SVR	Server	Prefix used for internal consumer groups or services.	common/constants.py103

Sources: rag/llm/embedding_model.py145-151 rag/svr/task_executor.py48-141 agent/canvas.py29 rag/llm/chat_model.py59-61 internal/router/router.go45-46 common/constants.py103