LightRAG: A More Efficient Solution than GraphRAG for RAG Systems?

Поділитися
Вставка
  • Опубліковано 27 січ 2025

КОМЕНТАРІ •

  • @kai_s1985
    @kai_s1985 3 місяці тому +14

    Thanks so much! You are the best channel when it comes to RAG. Please keep informing us about the latest advancements in this field.
    I have played with GraphRag. It can be expensive if you have tons of data, but considering how cheap GPT-4o and GTP-4o-mini have become, the price is not the biggest concern at least for my use case. I processes more than a thousand page document with GPT-4o and it cost me cents. The biggest problem with MS GraphRag is the inference latency. It is not very practical if you want to build a chatbot based on this. Also, it is less customizable in my recent experience. Hope LightRag is better in terms of accuracy, customizability and inference speed.

  • @srhino14
    @srhino14 2 місяці тому +1

    Thanks!

  • @CAPSLOCK_USER
    @CAPSLOCK_USER 3 місяці тому +3

    such a great channel, thanks for this guide, i was just about to implement a knowledge base!

  • @Imasoulinseoul
    @Imasoulinseoul 2 місяці тому

    Thank you heaps for the diagrams and explanations!

  • @johnkintree763
    @johnkintree763 3 місяці тому +11

    It looks like the pipeline is modular, and could work with different vector and graph databases.
    In the future, users could rate the responses, so highly rated responses could be stored and retrieved when similar queries are made by other users. Retrieving highly rated responses could improve overall system performance compared with generating responses for every query.
    Apply fast thinking instead of slow thinking when possible.

    • @KevinKreger
      @KevinKreger 2 місяці тому

      It is modular. For example, if you want to replace Textract with MinerU for doc/pdf parsing it would be very easy.

  • @AndrewNanton
    @AndrewNanton 3 місяці тому +3

    Pretty cool - if you do more with this, I'd love to see some experiements combining this with late chunking

    • @engineerprompt
      @engineerprompt  3 місяці тому +5

      good idea, when I get time, I want to implement a system that combines all these different approaches together.

    • @hiranga
      @hiranga 3 місяці тому +1

      @@engineerprompt Has anyone recreated this in LangGraph or LangChain JS ?

    • @MrAhsan99
      @MrAhsan99 Місяць тому

      @@engineerprompt looking forward!

  • @kenchang3456
    @kenchang3456 2 місяці тому +1

    Thanks for this. I was looking for a way to add to the KG without having to rebuild it. Crossing my fingers that this is it. And cheaper too 🙂

  • @FunDumb
    @FunDumb 3 місяці тому

    Definitely want to learn more about lite rag. Cost was a hindrance with regular rag. 😊

  • @amdenis
    @amdenis Місяць тому +1

    I started using a recursive variant of this for a bit, which evolved to a multi-LLM approach due to the need to optimize cost-performance efficiencies, but still leverage external inference time optimizations and multi-step sequencing and solving. I think most of these RAG and TAG mechanisms (light, long, standard GR, and the various fine and related tuning methods) will all continue to be superseded at an accelerating rate.
    The biggest problems I see in the industry from startups to universities and research groups is that the choices made and implementations used are often too brittle and subject to rip and replace requirements to be anywhere near cost-performance optimal in the long term, which for AI means even 1-2 years. So, better design patterns, tooling and implementation architectures are needed.

  • @dawid_dahl
    @dawid_dahl 3 місяці тому

    Thanks for the really great video, by the way.

  • @VaibhavPatil-rx7pc
    @VaibhavPatil-rx7pc 2 місяці тому

    Awesome 🎉

  • @marktahu2932
    @marktahu2932 Місяць тому +1

    I am assuming that eventually all publications will come with a link to their embeddings pre-prepared on some database somewhere associated with the publishers.

  • @yvescourtois
    @yvescourtois 3 місяці тому

    Great stuff, well done! I don't see any reason to use GraphRAG any more after that. I guess the technology will continue to surprise us, but the cost argument is powerful when you handle tons of data

  • @ahmadzaimhilmi
    @ahmadzaimhilmi 3 місяці тому +3

    It costs me $0.02 per 10 page pdf document using openai api. I think it's pretty decent.
    Future improvements should also include storing all these json files in a database.
    Also, have you figured out how to get the reference chunks used to generate the response to the query?

    • @penguinmagehello
      @penguinmagehello 3 місяці тому +1

      Mind sharing how/ which packages you use to split pdfs especially those with images?

    • @ahmadzaimhilmi
      @ahmadzaimhilmi 3 місяці тому

      @@penguinmagehello google marker pdf to markdown

    • @KevinKreger
      @KevinKreger 2 місяці тому

      @@penguinmagehello No images. The example on github is using textract for pdf text extraction, but you can use any other approach and save the images. You could add them to the graph by sending them to a VLM for a descriptive summary. That would go to the entity/relationship resolution step. Your chunking would have to respect the figure description (See figure 1, which blah blah) and the figure title.

  • @gunu3187
    @gunu3187 3 місяці тому +2

    Thanks for sharing , however this is another smart config changes resulting in low prompt tokens , Microsoft GraphRag uses higher overlap window ( 300:100 ) whereas here it is using (1200:100) which itself reduces number of prompt tokens used significantly. we should deep dive to understand if researchers have done something different to ensure much lower overlap window

  • @ai_dart
    @ai_dart 3 місяці тому +1

    Good Info

  • @mirkoappel
    @mirkoappel Місяць тому

    Thank you!

  • @trytry6569
    @trytry6569 3 місяці тому +6

    Please please please make a video in which we can use it with our local models.

    • @engineerprompt
      @engineerprompt  3 місяці тому +5

      on it :)

    • @trytry6569
      @trytry6569 3 місяці тому

      @@engineerprompt Thanks man🙌

    • @johnkintree763
      @johnkintree763 3 місяці тому +1

      ​@@engineerpromptThe Zamba2 family of models looks interesting for running locally with lower latency and more tokens/sec output.

  • @ashgtd
    @ashgtd 3 місяці тому

    great video thank you

  • @emirishak7609
    @emirishak7609 2 місяці тому +2

    Using LightRAG with gguf?

  • @KevinKreger
    @KevinKreger 2 місяці тому

    On the 'mixed' dataset, I don't think it had enough information on the various topics to create a comprehensive graph. I'm wondering if GraphRAG digs deeper or does some other technique with the graph meta-data like BM25.

  • @misterdesaster1021
    @misterdesaster1021 12 днів тому

    Thanks for the cool video. I have already set up a simple chatbot with this. Now I was wondering whether it is somehow possible for my answer to also contain the details of the sources from which the entities or the information for the answer originate. I haven't found anything about this in the docs yet. Do you or anyone else have any ideas?

  • @greglinklater6331
    @greglinklater6331 17 днів тому

    Thoughts about using late-chunking with Light RAG? Honestly I barely understand what's going on at a high level but conceptually is there anything obvious preventing late chunking from being applied in LightRAG?

  • @remusomega
    @remusomega 3 місяці тому +2

    Graphs solve the problem of chunk embeddings being de-contextualized.
    Late Chunking solves this problem. I think we need to re-consider the use cases for GraphRags.

  • @abhishekbose1081
    @abhishekbose1081 3 місяці тому +2

    Can I use this with Hugging face inference api or Vertexai api ? Actually I have access to this api only.

    • @yigidovic
      @yigidovic 2 місяці тому

      Use LiteLLM to standardize them to OpenAI API then configure BASE_URL and you are free!

    • @derekwang5982
      @derekwang5982 2 місяці тому

      I only processed 1 chunks for 2 hours, not 2 chunks being processed...not sure why

  • @isaatalay5320
    @isaatalay5320 День тому

    is there alternative easy interface based way for non code people ?

  • @Jvo_Rien
    @Jvo_Rien 3 місяці тому

    thank you :)

  • @guscastilloa
    @guscastilloa 3 місяці тому

    How do you record your screen so that it follows your cursos? It’s super useful to follow along specially when reading papers! Kudos on this amazing expose of this methodology!

    • @dylanmoraes990
      @dylanmoraes990 3 місяці тому

      It's a video editing app called screen studio on mac

  • @RickySupriyadi
    @RickySupriyadi 3 місяці тому +1

    wow so many RAG system this year already

    • @engineerprompt
      @engineerprompt  3 місяці тому +1

      Yup, its hard to track but nice to see the different ideas that are coming up.

  • @WaylonLu
    @WaylonLu 3 місяці тому

    how does integrate the vector dataabse?

  • @mclachan
    @mclachan 3 місяці тому

    Have you tested it on structured data?

  • @the_vheed1319
    @the_vheed1319 6 днів тому

    Hybird or hybrid?

  • @muhammadshafiqsafian6149
    @muhammadshafiqsafian6149 3 місяці тому

    how can i view the algo flowchart? a bit complex to understand

  • @AIWhale3
    @AIWhale3 3 місяці тому +1

    Is Lightrag supposed to be used with a vector database, a graph database or both?

    • @KevinKreger
      @KevinKreger 2 місяці тому

      Both (nano vector database) along with the graph (which I believe is Neo4J by default)

  • @dawid_dahl
    @dawid_dahl 3 місяці тому +19

    In a few years I suspect we will laugh at all these hacky RAG implementations as one will simply be able to dump everything into the context window and there will be some native mechanism to handle efficiency. What do you think?

    • @adegboyegaajenifuja1274
      @adegboyegaajenifuja1274 3 місяці тому

      Or you just grant access to your document repositories and you're good to go

    •  3 місяці тому +4

      Needle in haystack problem of big context Windows....

    • @BryantAvila
      @BryantAvila 3 місяці тому +5

      To dump 10K documents each composed of at least 100 pages seems it will always be unrealistic. RAG of some sort will still be needed.

    • @dawid_dahl
      @dawid_dahl 3 місяці тому +1

      @ Will be interesting to see how this comment ages over the coming years. (Or mine!)

    • @thunkin-ai
      @thunkin-ai 3 місяці тому +1

      the native mechanism might probably be a graphrag implementation

  • @syedsaifullahtarique
    @syedsaifullahtarique 3 місяці тому

    How to create LightRAG object inside dicken folder

  • @axelwehmeyer9599
    @axelwehmeyer9599 3 місяці тому

    cool.
    Is there an implementation for Groq-API like gpt_4o_mini_complete-API in LightRAG?
    How can i use a GUI-Chatbot for LightRAG, e.g. chainlit/streamlit/...?
    thx

    • @engineerprompt
      @engineerprompt  3 місяці тому

      I think there are a couple of PRs for other models. Not sure about the GUI

  • @akshatgandhi7958
    @akshatgandhi7958 3 місяці тому

    Thanks

  • @ingenierofelipeurreg
    @ingenierofelipeurreg 3 місяці тому

    Which of all RAG is cost effective and quality better?

    • @engineerprompt
      @engineerprompt  3 місяці тому +3

      Its a hard question to answer. It will really depend on the task at hands. If you think its just search/looking up information, may be a standard RAG. If there are relationships between objects/entities then may be a knowledge graph.

  • @viiskb
    @viiskb 2 місяці тому

    For the same book, graphRAG uses 1.1 million input tokens and 200k output tokens.

  • @ahmadzaimhilmi
    @ahmadzaimhilmi 3 місяці тому

    I like everything about lightrag. I wish we could see the reference chunk/document for the returned query too.
    Also, if there's a better way to query from a sql/nosql db instead of querying from drive.

  • @AhmedMagdy-ly3ng
    @AhmedMagdy-ly3ng 3 місяці тому

    Hey 👋
    I'm one of the most exciting fans of you and I have the opportunity to come to India I wish I can see you and have a conversation..
    Please 🥺🙏

  • @JoseAntonio-sn6sf
    @JoseAntonio-sn6sf 3 місяці тому

    nice video, I am just starting with RAGs, so sorry if my question is a little stupid, but if you spend 80k tokens running LightRAG, why the necessity to even implement a RAG when the book it self has 40k tokens? I mean wouldn't be easier to send the whole book to chatgpt?

    • @ravigurram9884
      @ravigurram9884 2 місяці тому +1

      Imagine large systems with terabytes of data.

    • @vap0rtranz
      @vap0rtranz 2 місяці тому

      > "I mean wouldn't be easier to send the whole book to chatgpt?" because this stuff is for knowledge graphs, not simple doc Q&A. The graphs create relationships *between* content. We used to do this in KBs by manually (and slowly) tagging and hyperinking stuff that we thought was related. Now models can create the relationships for us without having to hire 1,000s of librarians. The possibility to have a system trained to create meaningful relationships between content and dynamically pull that content on demand is mind blowing. Pre-trained models are already coming out.

  • @themax2go
    @themax2go 3 місяці тому +2

    so essentially they implemented triples / triplets (sciphi/triplex) 🤔