gmakstutis/LightRAG
1
0
Fork 0
Code Issues Pull requests Projects Releases Packages Wiki Activity Actions

cherry-pick 5c964267

Browse source
This commit is contained in:
Raphaël MANSUY 2025-12-04 19:19:24 +08:00
parent ff257ec5c1
commit e09569527f
1 changed files with 56 additions and 43 deletions
Show stats Download patch file Download diff file Expand all files Collapse all files

99
README.md
View file

@ -51,24 +51,24 @@
---
## 🎉 News
- [2025.11.05]🎯[New Feature]: Integrated **RAGAS for Evaluation** and **Langfuse for Tracing**. Updated the API to return retrieved contexts alongside query results to support context precision metrics.
- [2025.10.22]🎯Eliminate bottlenecks in processing **large-scale datasets**.
- [2025.09.15]🎯Significantly enhances KG extraction accuracy for **small LLMs** like Qwen3-30B-A3B.
- [2025.08.29]🎯**Reranker** is supported now , significantly boosting performance for mixed queries(Set as default query mode now).
- [2025.08.04]🎯**Document deletion** with KG regeneration to ensure query performance.
- [2025.06.16]🎯Our team has released [RAG-Anything](https://github.com/HKUDS/RAG-Anything) an All-in-One Multimodal RAG System for seamless text, image, table, and equation processing.
- [2025.06.05]🎯LightRAG now supports comprehensive multimodal data handling through [RAG-Anything](https://github.com/HKUDS/RAG-Anything) integration, enabling seamless document parsing and RAG capabilities across diverse formats including PDFs, images, Office documents, tables, and formulas. Please refer to the new [multimodal section](https://github.com/HKUDS/LightRAG/?tab=readme-ov-file#multimodal-document-processing-rag-anything-integration) for details.
- [2025.03.18]🎯LightRAG now supports citation functionality, enabling proper source attribution.
- [2025.02.12]🎯You can now use MongoDB as all in-one Storage.
- [2025.02.05]🎯Our team has released [VideoRAG](https://github.com/HKUDS/VideoRAG) understanding extremely long-context videos.
- [2025.01.13]🎯Our team has released [MiniRAG](https://github.com/HKUDS/MiniRAG) making RAG simpler with small models.
- [2025.01.06]🎯You can now use PostgreSQL as all in-one Storage.
- [2024.11.19]🎯A comprehensive guide to LightRAG is now available on [LearnOpenCV](https://learnopencv.com/lightrag). Many thanks to the blog author.
- [2024.11.09]🎯Introducing the LightRAG Webui, which allows you to insert, query, visualize LightRAG knowledge.
- [2024.11.04]🎯You can now [use Neo4J for Storage](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#using-neo4j-for-storage).
- [2024.10.18]🎯We've added a link to a [LightRAG Introduction Video](https://youtu.be/oageL-1I0GE). Thanks to the author!
- [2024.10.17]🎯We have created a [Discord channel](https://discord.gg/yF2MmDJyGJ)! Welcome to join for sharing and discussions! 🎉🎉
- [2024.10.16]🎯LightRAG now supports [Ollama models](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#quick-start)!
- [2025.11]🎯[New Feature]: Integrated **RAGAS for Evaluation** and **Langfuse for Tracing**. Updated the API to return retrieved contexts alongside query results to support context precision metrics.
- [2025.10]🎯[Scalability Enhancement]: Eliminated processing bottlenecks to support **Large-Scale Datasets Efficiently**.
- [2025.09]🎯[New Feature] Enhances knowledge graph extraction accuracy for **Open-Sourced LLMs** such as Qwen3-30B-A3B.
- [2025.08]🎯[New Feature] **Reranker** is now supported, significantly boosting performance for mixed queries (set as default query mode).
- [2025.08]🎯[New Feature] Added **Document Deletion** with automatic KG regeneration to ensure optimal query performance.
- [2025.06]🎯[New Release] Our team has released [RAG-Anything](https://github.com/HKUDS/RAG-Anything) — an **All-in-One Multimodal RAG** system for seamless processing of text, images, tables, and equations.
- [2025.06]🎯[New Feature] LightRAG now supports comprehensive multimodal data handling through [RAG-Anything](https://github.com/HKUDS/RAG-Anything) integration, enabling seamless document parsing and RAG capabilities across diverse formats including PDFs, images, Office documents, tables, and formulas. Please refer to the new [multimodal section](https://github.com/HKUDS/LightRAG/?tab=readme-ov-file#multimodal-document-processing-rag-anything-integration) for details.
- [2025.03]🎯[New Feature] LightRAG now supports citation functionality, enabling proper source attribution and enhanced document traceability.
- [2025.02]🎯[New Feature] You can now use MongoDB as an all-in-one storage solution for unified data management.
- [2025.02]🎯[New Release] Our team has released [VideoRAG](https://github.com/HKUDS/VideoRAG)-a RAG system for understanding extremely long-context videos
- [2025.01]🎯[New Release] Our team has released [MiniRAG](https://github.com/HKUDS/MiniRAG) making RAG simpler with small models.
- [2025.01]🎯You can now use PostgreSQL as an all-in-one storage solution for data management.
- [2024.11]🎯[New Resource] A comprehensive guide to LightRAG is now available on [LearnOpenCV](https://learnopencv.com/lightrag). — explore in-depth tutorials and best practices. Many thanks to the blog author for this excellent contribution!
- [2024.11]🎯[New Feature] Introducing the LightRAG WebUI — an interface that allows you to insert, query, and visualize LightRAG knowledge through an intuitive web-based dashboard.
- [2024.11]🎯[New Feature] You can now [use Neo4J for Storage](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#using-neo4j-for-storage)-enabling graph database support.
- [2024.10]🎯[New Feature] We've added a link to a [LightRAG Introduction Video](https://youtu.be/oageL-1I0GE). — a walkthrough of LightRAG's capabilities. Thanks to the author for this excellent contribution!
- [2024.10]🎯[New Channel] We have created a [Discord channel](https://discord.gg/yF2MmDJyGJ)!💬 Welcome to join our community for sharing, discussions, and collaboration! 🎉🎉
- [2024.10]🎯[New Feature] LightRAG now supports [Ollama models](https://github.com/HKUDS/LightRAG?tab=readme-ov-file#quick-start)!
<details>
<summary style="font-size: 1.4em; font-weight: bold; cursor: pointer; display: list-item;">
@ -214,7 +214,7 @@ For a streaming response implementation example, please see `examples/lightrag_o
**Note 2**: Only `lightrag_openai_demo.py` and `lightrag_openai_compatible_demo.py` are officially supported sample codes. Other sample files are community contributions that haven't undergone full testing and optimization.
## Programing with LightRAG Core
## Programming with LightRAG Core
> ⚠️ **If you would like to integrate LightRAG into your project, we recommend utilizing the REST API provided by the LightRAG Server**. LightRAG Core is typically intended for embedded applications or for researchers who wish to conduct studies and evaluations.
@ -313,7 +313,7 @@ A full list of LightRAG init parameters:
| **vector_db_storage_cls_kwargs** | `dict` | Additional parameters for vector database, like setting the threshold for nodes and relations retrieval | cosine_better_than_threshold: 0.2（default value changed by env var COSINE_THRESHOLD) |
| **enable_llm_cache** | `bool` | If `TRUE`, stores LLM results in cache; repeated prompts return cached responses | `TRUE` |
| **enable_llm_cache_for_entity_extract** | `bool` | If `TRUE`, stores LLM results in cache for entity extraction; Good for beginners to debug your application | `TRUE` |
| **addon_params** | `dict` | Additional parameters, e.g., `{"language": "Simplified Chinese", "entity_types": ["organization", "person", "location", "event"]}`: sets example limit, entiy/relation extraction output language | language: English` |
| **addon_params** | `dict` | Additional parameters, e.g., `{"language": "Simplified Chinese", "entity_types": ["organization", "person", "location", "event"]}`: sets example limit, entity/relation extraction output language | language: English` |
| **embedding_cache_config** | `dict` | Configuration for question-answer caching. Contains three parameters: `enabled`: Boolean value to enable/disable cache lookup functionality. When enabled, the system will check cached responses before generating new answers. `similarity_threshold`: Float value (0-1), similarity threshold. When a new question's similarity with a cached question exceeds this threshold, the cached answer will be returned directly without calling the LLM. `use_llm_check`: Boolean value to enable/disable LLM similarity verification. When enabled, LLM will be used as a secondary check to verify the similarity between questions before returning cached answers. | Default: `{"enabled": False, "similarity_threshold": 0.95, "use_llm_check": False}` |
</details>
@ -364,7 +364,7 @@ class QueryParam:
max_total_tokens: int = int(os.getenv("MAX_TOTAL_TOKENS", "30000"))
"""Maximum total tokens budget for the entire query context (entities + relations + chunks + system prompt)."""
# History mesages is only send to LLM for context, not used for retrieval
# History messages are only sent to LLM for context, not used for retrieval
conversation_history: list[dict[str, str]] = field(default_factory=list)
"""Stores past conversation history to maintain context.
Format: [{"role": "user/assistant", "content": "message"}].
@ -403,6 +403,11 @@ LightRAG requires the utilization of LLM and Embedding models to accomplish docu
* LightRAG also supports Open AI-like chat/embeddings APIs:
```python
import os
import numpy as np
from lightrag.utils import wrap_embedding_func_with_attrs
from lightrag.llm.openai import openai_complete_if_cache, openai_embed
async def llm_model_func(
prompt, system_prompt=None, history_messages=[], keyword_extraction=False, **kwargs
) -> str:
@ -416,8 +421,9 @@ async def llm_model_func(
**kwargs
)
@wrap_embedding_func_with_attrs(embedding_dim=4096, max_token_size=8192)
async def embedding_func(texts: list[str]) -> np.ndarray:
return await openai_embed(
return await openai_embed.func(
texts,
model="solar-embedding-1-large-query",
api_key=os.getenv("UPSTAGE_API_KEY"),
@ -428,16 +434,17 @@ async def initialize_rag():
rag = LightRAG(
working_dir=WORKING_DIR,
llm_model_func=llm_model_func,
embedding_func=EmbeddingFunc(
embedding_dim=4096,
func=embedding_func
)
embedding_func=embedding_func # Pass the decorated function directly
)
await rag.initialize_storages()
return rag
```
> **Important Note on Embedding Function Wrapping:**
>
> `EmbeddingFunc` cannot be nested. Functions that have been decorated with `@wrap_embedding_func_with_attrs` (such as `openai_embed`, `ollama_embed`, etc.) cannot be wrapped again using `EmbeddingFunc()`. This is why we call `xxx_embed.func` (the underlying unwrapped function) instead of `xxx_embed` directly when creating custom embedding functions.
</details>
<details>
@ -476,19 +483,20 @@ If you want to use Ollama models, you need to pull model you plan to use and emb
Then you only need to set LightRAG as follows:
```python
import numpy as np
from lightrag.utils import wrap_embedding_func_with_attrs
from lightrag.llm.ollama import ollama_model_complete, ollama_embed
@wrap_embedding_func_with_attrs(embedding_dim=768, max_token_size=8192)
async def embedding_func(texts: list[str]) -> np.ndarray:
return await ollama_embed.func(texts, embed_model="nomic-embed-text")
# Initialize LightRAG with Ollama model
rag = LightRAG(
working_dir=WORKING_DIR,
llm_model_func=ollama_model_complete, # Use Ollama model for text generation
llm_model_name='your_model_name', # Your model name
# Use Ollama embedding function
embedding_func=EmbeddingFunc(
embedding_dim=768,
func=lambda texts: ollama_embed(
texts,
embed_model="nomic-embed-text"
)
),
embedding_func=embedding_func, # Pass the decorated function directly
)
```
@ -527,22 +535,27 @@ ollama create -f Modelfile qwen2m
Tiy can use `llm_model_kwargs` param to configure ollama:
```python
import numpy as np
from lightrag.utils import wrap_embedding_func_with_attrs
from lightrag.llm.ollama import ollama_model_complete, ollama_embed
@wrap_embedding_func_with_attrs(embedding_dim=768, max_token_size=8192)
async def embedding_func(texts: list[str]) -> np.ndarray:
return await ollama_embed.func(texts, embed_model="nomic-embed-text")
rag = LightRAG(
working_dir=WORKING_DIR,
llm_model_func=ollama_model_complete, # Use Ollama model for text generation
llm_model_name='your_model_name', # Your model name
llm_model_kwargs={"options": {"num_ctx": 32768}},
# Use Ollama embedding function
embedding_func=EmbeddingFunc(
embedding_dim=768,
func=lambda texts: ollama_embed(
texts,
embed_model="nomic-embed-text"
)
),
embedding_func=embedding_func, # Pass the decorated function directly
)
```
> **Important Note on Embedding Function Wrapping:**
>
> `EmbeddingFunc` cannot be nested. Functions that have been decorated with `@wrap_embedding_func_with_attrs` (such as `openai_embed`, `ollama_embed`, etc.) cannot be wrapped again using `EmbeddingFunc()`. This is why we call `xxx_embed.func` (the underlying unwrapped function) instead of `xxx_embed` directly when creating custom embedding functions.
* **Low RAM GPUs**
In order to run this experiment on low RAM GPU you should select small model and tune context window (increasing context increase memory consumption). For example, running this ollama example on repurposed mining GPU with 6Gb of RAM required to set context size to 26k while using `gemma2:2b`. It was able to find 197 entities and 19 relations on `book.txt`.
@ -1555,7 +1568,7 @@ Langfuse provides a drop-in replacement for the OpenAI client that automatically
pip install lightrag-hku
pip install lightrag-hku[observability]
# Or install from souce code with debug mode enabled
# Or install from source code with debug mode enabled
pip install -e .
pip install -e ".[observability]"
```