LightRAG

(★ 39,056)

[EMNLP2025] LightRAG: Simple and Fast Retrieval-Augmented Generation

파일 탐색기

최종 버전 다운로드 (.zip)

- - session-start.sh
- settings.json
- 01-basic.md
- - bug_report.yml
  - config.yml
  - feature_request.yml
  - question.yml
- - copilot-setup-steps.yml
  - docker-build-lite.yml
  - docker-build-manual.yml
  - docker-publish.yml
  - linting.yaml
  - pg-smoke.yml
  - pypi-publish.yml
  - stale.yaml
  - tests.yml
- CONTRIBUTING.md
- dependabot.yml
- pull_request_template.md
- LiteWrite.png
- logo.png
- - image-20250323122538997.png
  - image-20250323122754387.png
  - image-20250323123011220.png
  - image-20250323194750379.png
- AppleContainerSetup.md
- AsymmetricEmbedding.md
- DockerDeployment.md
- FileProcessingPipeline-zh.md
- FileProcessingPipeline.md
- FrontendBuildGuide.md
- InteractiveSetup.md
- LightRAG-API-Server-zh.md
- LightRAG-API-Server.md
- LightRAGSidecarFormat-zh.md
- LightRAGSidecarFormat.md
- LLMProviderOptions.md
- MilvusConfigurationGuide.md
- MultiSiteDeployment.md
- OfflineDeployment.md
- ParagraphSemanticChunking-zh.md
- ParagraphSemanticChunking.md
- ParserDebugCLI-zh.md
- ParserDebugCLI.md
- ParserServiceDeployment-zh.md
- ParserServiceDeployment.md
- ProgramingWithCore.md
- Reproduce.md
- RoleSpecificLLMConfiguration-zh.md
- RoleSpecificLLMConfiguration.md
- ThirdPartyParser-zh.md
- ThirdPartyParser.md
- UV_LOCK_GUIDE.md
- - copy_llm_cache_to_another_storage.py
  - lightrag_bedrock_demo.py
  - lightrag_cloudflare_demo.py
  - lightrag_embedding_prefixes.py
  - lightrag_hf_demo.py
  - lightrag_llamaindex_direct_demo.py
  - lightrag_llamaindex_litellm_demo.py
  - lightrag_llamaindex_litellm_opik_demo.py
  - lightrag_lmdeploy_demo.py
  - lightrag_nvidia_demo.py
  - lightrag_openai_neo4j_milvus_redis_demo.py
- generate_query.py
- graph_visual_with_html.py
- graph_visual_with_neo4j.py
- graph_visual_with_opensearch.py
- insert_custom_kg.py
- lightrag_ag2_multiagent_demo.py
- lightrag_azure_openai_demo.py
- lightrag_gemini_demo.py
- lightrag_gemini_postgres_demo.py
- lightrag_gemini_workspace_demo.py
- lightrag_ollama_demo.py
- lightrag_openai_compatible_demo.py
- lightrag_openai_demo.py
- lightrag_openai_mongodb_graph_demo.py
- lightrag_openai_opensearch_graph_demo.py
- lightrag_vllm_demo.py
- milvus_kwargs_configuration_demo.py
- opensearch_storage_demo.py
- rerank_example.py
- - 00-config.sh
  - 01-prepare.sh
  - 02-install-database.sh
  - 03-uninstall-database.sh
  - 04-cleanup.sh
  - install-kubeblocks.sh
  - K8S-DB-README.md
  - uninstall-kubeblocks.sh
- - .helmignore
  - Chart.yaml
  - values.yaml
- install_lightrag.sh
- install_lightrag_dev.sh
- K8S-README-zh.md
- K8S-README.md
- uninstall_lightrag.sh
- uninstall_lightrag_dev.sh
- - .env.aoi.example
  - .gitignore
  - __init__.py
  - admission.py
  - admission_middleware.py
  - asgi_helpers.py
  - auth.py
  - body_limit_middleware.py
  - config.py
  - gunicorn_config.py
  - gunicorn_worker.py
  - input_limits.py
  - lightrag_server.py
  - login_rate_limit.py
  - passwords.py
  - run_with_gunicorn.py
  - runtime_validation.py
  - utils_api.py
- - __init__.py
  - paragraph_semantic.py
  - recursive_character.py
  - semantic_vector.py
  - token_size.py
- - __init__.py
  - eval_rag_quality.py
  - offline_retrieval_check.py
  - README_EVALUASTION_RAGAS.md
  - sample_dataset.json
  - sample_retrieval_oracle.json
- - __init__.py
  - factory.py
  - faiss_impl.py
  - json_doc_status_impl.py
  - json_kv_impl.py
  - memgraph_impl.py
  - milvus_impl.py
  - mongo_impl.py
  - nano_vector_db_impl.py
  - neo4j_impl.py
  - networkx_impl.py
  - noop_vector_db_impl.py
  - opensearch_impl.py
  - pgtable_impl.py
  - pipeline_ingress.py
  - postgres_impl.py
  - qdrant_impl.py
  - redis_impl.py
  - scan_job_store.py
  - shared_storage.py
- - __init__.py
  - _vision_utils.py
  - anthropic.py
  - azure_openai.py
  - bedrock.py
  - binding_options.py
  - gemini.py
  - hf.py
  - jina.py
  - llama_index_impl.py
  - lmdeploy.py
  - lollms.py
  - nvidia_openai.py
  - ollama.py
  - openai.py
  - voyageai.py
  - zhipu.py
- - __init__.py
  - _html_table.py
  - _markdown.py
  - base.py
  - cli.py
  - debug.py
  - exceptions.py
  - llm_bridge.py
  - native_base.py
  - native_dispatch.py
  - noop.py
  - param_schema.py
  - plugins.py
  - registry.py
  - routing.py
- - __init__.py
  - backfill.py
  - ir.py
  - placeholders.py
  - writer.py
- - __init__.py
  - check_initialization.py
  - clean_llm_query_cache.py
  - download_cache.py
  - hash_password.py
  - kg_integrity_repair.py
  - migrate_graph_storage.py
  - migrate_llm_cache.py
  - README_CLEAN_LLM_QUERY_CACHE.md
  - README_KG_INTEGRITY_REPAIR.md
  - README_MIGRATE_GRAPH_STORAGE.md
  - README_MIGRATE_LLM_CACHE.md
  - README_REBUILD_VDB.md
  - README_SOURCE_CONFLICT_REPAIR.md
  - rebuild_vdb.py
  - source_conflict_repair.py
- __init__.py
- _version.py
- addon_params.py
- base.py
- chunk_schema.py
- constants.py
- exceptions.py
- file_atomic.py
- lightrag.py
- llm_roles.py
- multimodal_context.py
- namespace.py
- operate.py
- pipeline.py
- pipeline_metrics.py
- prompt.py
- prompt_multimodal.py
- rerank.py
- storage_migrations.py
- table_markup.py
- types.py
- utils.py
- utils_graph.py
- utils_pipeline.py
- - favicon.png
  - logo.svg
- - App.tsx
  - AppRouter.tsx
  - i18n.ts
  - index.css
  - main.tsx
  - vite-env.d.ts
- .env.development
- .gitignore
- .prettierrc.json
- bun.lock
- components.json
- env.development.smaple
- env.local.sample
- eslint.config.js
- index.html
- package.json
- tailwind.config.js
- tsconfig.json
- vite.config.ts
- WEBUI-README.md
- - entity_type_prompt.sample.yml
- UserCustomizePrompts.md
- b2aaf634151b4706892693ffb43d9093.png
- iShot_2025-03-23_12.40.08.png
- batch_eval.py
- Step_0.py
- Step_1.py
- Step_1_openai_compatible.py
- Step_2.py
- Step_3.py
- Step_3_openai_compatible.py
- - set_version.py
- - apple-container.sh
  - setup.sh
- regen_native_docx_golden.py
- test.sh
- - __init__.py
  - conftest.py
  - test_admission_middleware.py
  - test_api_docs_toggle.py
  - test_body_limit_middleware.py
  - test_cors.py
  - test_error_message_sanitization.py
  - test_gunicorn_worker_orphan_exit.py
  - test_health_auth.py
  - test_health_scheduling_view.py
  - test_lightrag_ollama_chat.py
  - test_ollama_query_mode.py
  - test_path_prefixes.py
  - test_recovery_fence_visibility.py
- - __init__.py
  - test_chunk_options_persistence.py
  - test_chunker_recursive_character.py
  - test_chunker_semantic_vector.py
  - test_chunking.py
  - test_chunking_raw_lightrag_parity.py
  - test_hard_fallback_span_projection.py
  - test_overlap_validation.py
  - test_paragraph_semantic_drop_references.py
  - test_paragraph_semantic_merge_and_fallback.py
  - test_paragraph_semantic_split_long_block.py
  - test_paragraph_semantic_table_split.py
  - test_paragraph_semantic_title_block.py
  - test_recursive_character_bounds.py
  - test_rerank_chunking.py
  - test_sidecar_backfill_integration.py
  - test_token_window_source_span.py
- - __init__.py
  - test_evaluation_offline_retrieval_check.py
- - __init__.py
  - test_chunk_results_ordering.py
  - test_chunk_tracking_presence_ingestion.py
  - test_edge_vdb_none_guard.py
  - test_edge_weight_reprocess.py
  - test_entity_extraction_stability.py
  - test_extract_entities.py
  - test_extract_entities_empty_chunks.py
  - test_fix_tuple_delimiter_free_text.py
  - test_json_extraction_recovery.py
  - test_keyword_extraction_drivers.py
  - test_keyword_parsing.py
  - test_keyword_prompt_template.py
  - test_kg_recovery_primitives.py
  - test_merge_description_dedup.py
  - test_merge_entities_chunk_tracking.py
  - test_merge_entities_relation_weight.py
  - test_pipeline_log_delock.py
  - test_rebuild_history_scope.py
  - test_rebuild_synthesis.py
  - test_relation_evidence_floor_repair.py
  - test_relation_numeric_description_weight.py
  - test_rename_tracking_migration_ordering.py
  - test_reserved_entity_type_rejection.py
  - test_summary_map_reduce_perf.py
  - test_summary_sanitization.py
  - test_truncate_vdb_content.py
  - test_truncation_reporting.py
  - test_vdb_content_truncation_coverage.py
  - test_write_ahead_indexes.py
- - __init__.py
  - conftest_scheduling_backends.py
  - test_batch_graph_operations.py
  - test_bounded_pipeline_history.py
  - test_committed_background_task.py
  - test_degree_return_type.py
  - test_doc_status_scheduling_contract.py
  - test_doc_status_strict_parse_branches.py
  - test_file_atomic.py
  - test_get_all_update_flags_status.py
  - test_global_concurrency_slots.py
  - test_graph_storage.py
  - test_keyed_holder_table.py
  - test_keyed_lock_async_registry.py
  - test_keyed_lock_cancel_safety.py
  - test_keyed_lock_dead_worker_recovery.py
  - test_pipeline_ingress.py
  - test_pipeline_status_logger.py
  - test_reservation_architecture.py
  - test_reservation_dead_process_recovery.py
  - test_reservation_primitives.py
  - test_reserved_background_task.py
  - test_scan_job_store.py
  - test_scheduling_backend_fixture.py
  - test_scheduling_base_defaults.py
  - test_scheduling_page_plans.py
  - test_shared_storage_rpc_counts.py
  - test_unified_lock_safety.py
- - __init__.py
  - test_apply_rerank_result_validation.py
  - test_asymmetric_embedding.py
  - test_batch_embeddings.py
  - test_bounded_submit.py
  - test_bypass_truncated_response.py
  - test_dimension_mismatch.py
  - test_global_concurrency_queue.py
  - test_llm_binding_image_payload.py
  - test_llm_cache_identity.py
  - test_llm_role_runtime.py
  - test_no_model_suffix_safety.py
  - test_process_chunks_unified_two_stage_truncation.py
  - test_query_cache_conversation_history.py
  - test_query_truncated_response_not_cached.py
  - test_remove_think_tags.py
  - test_runtime_target_validation.py
  - test_token_executor.py
  - test_tokenizer_contract.py
  - test_tokenizer_split_truncate_contract.py
  - test_utils_llm_cache.py
  - test_vlm_cache_key.py
  - test_vlm_image_inputs.py
  - test_vlm_json_escape_repair.py
- - __init__.py
  - test_chunk_r_separators_env.py
  - test_chunk_size_empty_env.py
  - test_chunk_v_buffer_empty_env.py
  - test_engine_param_cache_signature.py
  - test_engine_params.py
  - test_hint_params.py
  - test_legacy_extractors.py
  - test_legacy_parser.py
  - test_llm_bridge.py
  - test_markdown.py
  - test_parse_native_lightrag_e2e.py
  - test_parser_cli.py
  - test_parser_exceptions.py
  - test_plugins.py
  - test_registry.py
  - test_smart_heading_global_default.py
- - __init__.py
  - conftest.py
  - test_admission_capacity.py
  - test_bounded_scheduling.py
  - test_chunk_options_regex_scrub.py
  - test_chunking_executor_bounds.py
  - test_chunking_func_contract.py
  - test_chunking_off_event_loop.py
  - test_chunking_snapshot_replay.py
  - test_content_hash_dedup.py
  - test_content_hash_normalization.py
  - test_custom_chunk_patch.py
  - test_custom_chunk_rollback.py
  - test_custom_chunking_selector.py
  - test_custom_kg_self_loop_rejection.py
  - test_delete_fail_closed.py
  - test_demotion_survives_the_pipeline.py
  - test_doc_status_chunk_preservation.py
  - test_doc_status_reset_metadata.py
  - test_document_file_path_normalization.py
  - test_embedding_chunk_overlap_config.py
  - test_empty_chunks_document.py
  - test_enqueue_basename_strict.py
  - test_enqueue_created_at.py
  - test_enqueue_ingress_exclusion.py
  - test_enqueue_smart_heading_seed.py
  - test_fault_injection_matrix.py
  - test_graph_keyed_locks.py
  - test_insert_done_error_handling.py
  - test_manual_drain_progress.py
  - test_manual_exclusive_reset.py
  - test_multimodal_content_truncation.py
  - test_multimodal_path_containment.py
  - test_multimodal_surrounding_context.py
  - test_parse_engine_early_persist.py
  - test_parse_failure_doc_status_fields.py
  - test_parse_warnings_split_persist.py
  - test_parse_worker_content_sanitized.py
  - test_persist_parsed_control_chars.py
  - test_pipeline_analyze_multimodal.py
  - test_pipeline_cancellation.py
  - test_pipeline_content_reread.py
  - test_pipeline_failed_retry_semantics.py
  - test_pipeline_ingress_exit.py
  - test_pipeline_ingress_feed.py
  - test_pipeline_internal_abort.py
  - test_pipeline_metrics.py
  - test_pipeline_orchestrator_get_by_id_failure.py
  - test_pipeline_release_closure.py
  - test_processed_after_flush.py
  - test_purge_fail_closed.py
  - test_purge_primitive.py
  - test_scheduler_memory_bounded.py
  - test_scheduler_memory_bounded_e2e.py
  - test_strict_capability_gate.py
  - test_tolerant_load_json_dict.py
  - test_truncation_doc_status_metadata.py
  - test_validator_strict_content_probe.py
  - test_vlm_empty_int_env.py
- - __init__.py
  - _helpers.py
  - test_apple_container.py
  - test_collect.py
  - test_env.py
  - test_generate.py
  - test_misc.py
  - test_validate.py
- - __init__.py
  - test_backfill.py
  - test_sidecar_uri.py
  - test_writer.py
- - __init__.py
  - conftest.py
  - test_clean_llm_query_cache_modes.py
  - test_migrate_graph_storage.py
  - test_migrate_graph_storage_networkx.py
  - test_migrate_llm_cache.py
  - test_rebuild_vdb.py
  - test_source_conflict_repair_cli.py
- - __init__.py
  - test_compute_args_hash.py
  - test_compute_incremental_chunk_ids.py
  - test_graph_attribute_validator.py
  - test_has_chunk_tracking_row.py
  - test_load_json.py
  - test_merge_source_ids.py
  - test_pick_by_vector_similarity.py
  - test_safe_log_value.py
  - test_truncation_tally.py
- - __init__.py
  - README_WORKSPACE_ISOLATION_TESTS.md
  - test_workspace_isolation.py
  - test_workspace_migration_isolation.py
  - test_workspace_path_validation.py
  - test_workspace_sanitization.py
- __init__.py
- conftest.py
- test_cosine_similarity.py
- test_cosine_similarity_nonfinite.py
- test_cosine_threshold_env.py
- test_create_prefixed_exception.py
- test_doc_processing_status_from_stored.py
- test_empty_int_env_field_defaults.py
- test_get_env_value_nonfinite.py
- test_get_env_value_padded_bool.py
- test_operate_vector_context.py
- test_parse_optional_float.py
- test_queryparam_empty_int_env.py
- test_rerank_null_chunk_index_score.py
- test_split_text_by_token_limit.py
- test_strip_control_characters.py
- test_sync_wrapper_guard.py
- test_weighted_polling_nonpositive_max.py
.dockerignore
.git-blame-ignore-revs
.gitattributes
.gitignore
.pre-commit-config.yaml
AGENTS.md
CLAUDE.md
config.ini.example
docker-build-push.sh
docker-compose-full.yml
docker-compose.podman.yml
docker-compose.yml
docker-entrypoint.sh
Dockerfile
Dockerfile.lite
Dockerfile.postgres
env.docker-compose-full
env.example
LICENSE
lightrag.service.example
Makefile
MANIFEST.in
pyproject.toml
README-ja.md
README-zh.md
README.md
requirements-offline-llm.txt
requirements-offline-smart-heading.txt
requirements-offline-storage.txt
requirements-offline.txt
SECURITY.md
setup.py
uv.lock

# 설치 가이드

1. 코드 내려받기

git clone https://github.com/HKUDS/LightRAG

깃허브에서 프로젝트 코드 전체를 내 컴퓨터로 내려받습니다.

cd LightRAG

방금 내려받은 프로젝트 폴더 안으로 이동합니다.

2. 공식 설치 스크립트

쉬움 추천

사전 준비물

APT (Debian/Ubuntu 계열) Debian/Ubuntu 계열 리눅스에 기본 내장된 패키지 매니저입니다.
Homebrew macOS/Linux용 패키지 매니저입니다.

sudo apt-get install -y libcairo2

APT 패키지 저장소에서 바로 설치합니다 (Debian/Ubuntu 계열).

brew install cairo

Homebrew로 이미 빌드된 패키지를 바로 설치합니다. 소스 빌드가 필요 없습니다.

✅ 설치 후 새 터미널을 열고, 프로그램의 버전 확인 명령(예: --version)으로 정상 설치됐는지 확인하세요.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

3. Docker

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Docker Desktop 컨테이너를 빌드하고 실행하려면 필요합니다. 설치 후 실행해서 백그라운드에 켜두세요.

docker compose up

정의된 모든 컨테이너(서버, DB 등)를 한 번에 빌드하고 백그라운드에서 실행합니다.

✅ 터미널에 docker compose ps 를 입력해 컨테이너들이 Up 상태인지 확인하세요. README에 포트 번호가 적혀있다면 브라우저에서 http://localhost:포트번호 로 접속해보세요.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

4. Node.js

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Node.js npm 명령어를 쓰려면 Node.js가 설치되어 있어야 합니다. LTS(장기지원) 버전을 추천합니다.

cd lightrag_webui

이 프로젝트의 관련 파일이 하위 폴더 안에 있어서, 먼저 그 폴더로 이동합니다.

npm install

package.json에 명시된 라이브러리들을 내려받아 설치합니다.

npm start

개발/실행 서버를 켭니다.

✅ 명령어 실행 후 터미널에 나타나는 주소(보통 http://localhost:3000 형태)를 브라우저에서 열어보세요.

5. Python

쉬움

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Python 3 설치 시 'Add Python to PATH' 옵션을 꼭 체크하세요 (Windows).

uv pip install lightrag-hku

requirements.txt 등에 명시된 파이썬 라이브러리를 설치합니다.

✅ 에러 메시지 없이 실행되고 터미널에 안내 문구가 출력되면 정상입니다.

이 레포의 README에 적힌 실제 명령어를 그대로 가져왔습니다.

6. Make

보통

사전 준비물

Git GitHub에서 프로젝트 코드를 내려받으려면 필요합니다.
Make Linux/macOS는 보통 기본 설치되어 있습니다. Windows는 별도 설치(예: MSYS2, WSL)가 필요합니다.

make dev