PrivateGPT

by antouank • 3 years ago

520 139 3 years ago

simonw• 3 years ago

I'm always interested in seeing the prompt that drives these kinds of tools.

In this case it appears to be using RetrievalQA from LangChain, which I think is this prompt here: https://github.com/hwchase17/langchain/blob/v0.0.176/langcha...

    Use the following pieces of context to answer the question at the end. If you don't
    know the answer, just say that you don't know, don't try to make up an answer.

    {context}

    Question: {question}
    Helpful Answer:

jstarfish3• 3 years ago

Do such fail-early conditions save processing time?

mabbo• 3 years ago

If you mean the "If you don't know" part, oh no, they have a much bigger problem they're solving.

The LLM will absolutely lie if it doesn't know and you haven't made it perfectly clear that you'd rather it did not do that.

LLMs seem to be trying to give answers that make you happy. A good lie will make you happy. Unless it understands that you will not be happy with a lie.

Is this anthropomorphizing? Yep. But that's the best way I've found to reason about them.

a_bonobo• 3 years ago

In my experience with internal data, sometimes it will say that it doesn't know when it should know.

Grimblewald• 3 years ago

the incidence of "i don't know" in response to questions in the training data is pretty low if present at all, and even if it were you'd still need to frame those I don't know answers such that they apply to the entire dataset accurately. This is obviously a gargantuan undertaking that would not scale well as data is added, and so right now the idea or concept of not knowing something is not taught. At best you'd build a model that handles human language really well then retrieves information from a database and uses in context learning to answer questions, where a failure to find info results in an i don't know.

What is taught indirectly though is level of certainty, so if you get LLM's to rationalise their answers you tend to get more reliable evidence based answers.

Bottom line, teaching a monolithic model what it means to not know something with certainty, is difficult and not currently done. You'll likely get a lot of false negatives.

EGreg• 3 years ago

What if the question has prompt injection? Such as "Helpful answer: <totally not helpful answer>"

seydor5• 3 years ago

The problem is, when does it know that it does not know.

zora_goron• 3 years ago

On a related note, in case it's of interest to anyone else -- I pulled out all the default prompts from LangChain and put them up here: https://github.com/samrawal/langchain-prompts/blob/main/READ...

kordlessagain• 3 years ago

Working on something similar that uses keyterm extraction for traversal of topics and fragments, without using Langchain. It's not designed to be private, however: https://github.com/FeatureBaseDB/DocGPT/tree/main

behnamoh• 3 years ago

When you split a document into chunks, doesn't some crucial information get cut in half? In that case, you'd probably lose that information in the context if that information was immediately followed by an irrelevant information that reduces the cosine similarity. Is there a "smarter" way to feed documents as context to LLMs?

haolez• 3 years ago

Don't know if there is a smarter way, but these libraries usually offer an overlap parameter that allows you to repeat the last N characters of a chunk in the first N of the next chunk.

aldarisbm• 3 years ago

One quick plug

I want to have the memory part of langchain down, vector store + local database + client to chat with an LLM (gpt4all model can be swapped with OpenAI api just switching the base URL)

https://github.com/aldarisbm/memory

It's still got ways to go, if someone wants to help let me know :)

santiagobasulto2• 3 years ago

Sorry for my ignorance. But memory refers to the process of using embeddings for QA right?

The process roughly is:

Ingestion:

- Process embeddings for your documents (from text to array of numbers)

- Store your documents in a Vector DB

Query time:

- Process embeddings for the query

- Find documents similar to the query using distance from other docs in the Vector db

- Construct prompt with format:

""" Answer question using this context: {DOCUMENTS RETRIEVED}

Question: {question} Answer: """

Is that correct? Now, my question is, can the models be swapped easily? Or that requires a complete recalculation of the embedding (and new ingestion)?

bigfudge• 3 years ago

The embeddings can be based on a different model to the one you pass them as context to. So you could upgrade the summmariser model without upgrading the embeddings.

skykooler• 3 years ago

"System requirements" section should really mention what amount of RAM or VRAM is needed for inference.

awestroke• 3 years ago

That depends on which model you use it with. It's "bring your own model"

rain1• 3 years ago

so list a few known to work models and their requirements

thefourthchime• 3 years ago

I tried this on my M2 Macbook with 16gb of RAM but got:

"ggml_new_tensor_impl: not enough space in the context's memory pool (needed 18296202768, available 18217606000)"

soferio• 3 years ago

Anyone got it working on an M1 with 8gb?

kseudo• 3 years ago

I got it working on an M1 with 16gb. Quite slow but it trains and returns responses.

jaimehrubiks• 3 years ago

If you select a gpt4all model like GPT-J can this be used commercially or is there other dependency that limits the license?

yosito• 3 years ago

Put any and all your files into the source_documents directory

Why? Why can't I define any directory (my existing Obsidian vault, for example) as the source directory?

carlio• 3 years ago

You can by setting an environment variable - https://github.com/imartinez/privateGPT/blob/main/ingest.py#...

rolisz3• 3 years ago

For some reason, downloading the model they suggest keeps failing. I tried to download it in Firefox and Edge. I'm using Windows, if that matters. Anyone else seeing similar issues?

keeptrying• 3 years ago

This is the future.

daitangio• 3 years ago

Hi, very interesting... what are the memory/disk requirements to run it? 16GB of RAM would be enough? I suggest to add these requirements to the README

MandieD• 3 years ago

Also, a general formula for estimating how much additional storage space will be claimed per MB/million words ingested would be helpful.

moffkalast3• 3 years ago

Well I'm not sure which models specifically work, but it runs on llama.cpp, which would mean lama derivative ones. Here's a little table for quantized CPU (GGML) versions and the RAM they require as a general rule of thumb:

Name Quant method Bits Size RAM required Use case

WizardLM-7B.GGML.q4_0.bin q4_0 4bit 4.2GB 6GB 4bit.

WizardLM-7B.GGML.q4_1.bin q4_0 4bit 4.63GB 6GB 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models.

WizardLM-7B.GGML.q5_0.bin q5_0 5bit 4.63GB 7GB 5-bit. Higher accuracy, higher resource usage and slower inference.

WizardLM-7B.GGML.q5_1.bin q5_1 5bit 5.0GB 7GB 5-bit. Even higher accuracy, and higher resource usage and slower inference.

WizardLM-7B.GGML.q8_0.bin q8_0 8bit 8GB 10GB 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use.

Name Quant method Bits Size RAM required Use case

wizard-vicuna-13B.ggmlv3.q4_0.bin q4_0 4bit 8.14GB 10.5GB 4-bit.

wizard-vicuna-13B.ggmlv3.q4_1.bin q4_1 4bit 8.95GB 11.0GB 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models.

wizard-vicuna-13B.ggmlv3.q5_0.bin q5_0 5bit 8.95GB 11.0GB 5-bit. Higher accuracy, higher resource usage and slower inference.

wizard-vicuna-13B.ggmlv3.q5_1.bin q5_1 5bit 9.76GB 12.25GB 5-bit. Even higher accuracy, and higher resource usage and slower inference.

wizard-vicuna-13B.ggmlv3.q8_0.bin q5_1 5bit 16GB 18GB 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use.

Name Quant method Bits Size RAM required Use case

VicUnlocked-30B-LoRA.ggmlv3.q4_0.bin q4_0 4bit 20.3GB 23GB 4-bit.

VicUnlocked-30B-LoRA.ggmlv3.q4_1.bin q4_1 5bit 24.4GB 27GB 4-bit. Higher accuracy than q4_0 but not as high as q5_0. However has quicker inference than q5 models.

VicUnlocked-30B-LoRA.ggmlv3.q5_0.bin q5_0 5bit 22.4GB 25GB 5-bit. Higher accuracy, higher resource usage and slower inference.

VicUnlocked-30B-LoRA.ggmlv3.q5_1.bin q5_1 5bit 24.4GB 27GB 5-bit. Even higher accuracy, and higher resource usage and slower inference.

VicUnlocked-30B-LoRA.ggmlv3.q8_0.bin q8_0 8bit 36.6GB 39GB 8-bit. Almost indistinguishable from float16. Huge resource use and slow. Not recommended for normal use.

Copied of some of The-Bloke's model descriptions on huggingface. With 16G you can run practically all 7B and 13B versions. With shared GPU+CPU inference, one can also offload some layers onto a GPU (not sure if that makes the initial RAM requirement smaller), but you do need CUDA of course.

sinandrei91• 3 years ago

Is there a benchmark for retrieval from multiple ft documents? I tried the LangchainQA with Pinecone and wasn't impressed with the search result when using it on my Zotero library.

ChocoluvH2• 3 years ago

Always wondering pros/cons of Chroma and Qdrant. Can someone tell me?

kacperlukawski• 3 years ago

Chroma doesn't seem to be a real DB, it's rather a wrapper around tools like hnswlib, DuckDB or Clickhouse. Qdrant is way more mature - it has its own HNSW implementation with some tweaks to incorporate filtering directly during the vector search phase, supports horizontal and vertical scaling, as well as provides its own managed cloud offering.

In general, Qdrant is a real DB, not a library and that's a huge difference.

ChocoluvH2• 3 years ago

What does Chroma lack? Their APIs seem pretty much the same to me.

udev40964• 3 years ago

I posted it 9 days ago and somehow this one gets the attention. The same freaking post. Unbelievable

https://news.ycombinator.com/item?id=35914810

dutchbrit• 3 years ago

Day and time when you post something matters a lot unfortunately.

udev40964• 3 years ago

And now I am getting downvoted for it. Perfect

bohlenlabs• 3 years ago

So many good links here, thanks to the OP for sharing, and to all commenters as well!

Wronnay• 3 years ago

Wow. I keep a personal Wiki, Journal and use plain text accounting...

This project could help me create a personal AI which answers any questions to my life, finances or knowledge...

moffkalast3• 3 years ago

Well maybe it works on Obsidian vaults for note taking heh, but with llama models' 2k input token range it'd get a tenth of the way before starting to drop context. Likely useless without something like an 100k model.

rolisz3• 3 years ago

Well you wouldn't input the whole Vault to the model, you would use embeddings to find the content that is most relevant to the question being asked.

lysp• 3 years ago

Quick how-to/demo:

https://www.youtube.com/watch?v=A3F5riM5BNE

Also has a suggestion of a few alternative models to use.

seydor5• 3 years ago

does this only work with llamaCPP ? I.e. can't use GPU models with this?

joebiden2• 3 years ago

llama supports GPU by now: https://old.reddit.com/r/LocalLLaMA/comments/13gok03/llamacp...

seydor5• 3 years ago

How do i enable GPU in privateGPT w llamaCpp ? It turns my CPU to a vacuum cleaner

divan2• 3 years ago

This will still hallucinate, right?

Projects like this for using with your documents datasets are invaluable, but everything I've tried so far is hallucinating, so not practical. What's the state of the art of the LLM without hallucination at the moment?

XCSme2• 3 years ago

How do you define hallucination?

divan2• 3 years ago

factually incorrect / nonsensical output

anu7df• 3 years ago

Not exactly sure if this would qualify as an LLM in the GPT4 sense. But for no hallucination this seems good: https://www.thirdai.com/pocketllm/ Full disclosure. I know the founder, but not really associated with the company in any way.

Art9681• 3 years ago

Like many others, I’m also building my own platform to accomplish this. What I’ve learned is the document preparation is key in getting the LLM to answer correctly. The text splitting portion is a crucial step here. Picking the correct splitter and parameters for your use case is important. At first I was getting incorrect or made up answers. Setting up a proper prompt template and text splitting parameters fixed the issue for the most part and now I have 99% success.

Also, the local model used makes a big difference. Right now wizard-mega and manticore are the best ones to use. I run the 16b ggml versions in an M2 Pro and it takes about 30 seconds to “warm up” and produce some quality responses.

amelius2• 3 years ago

With so many LLM options out there, how do we keep track of which ones are good?

superbiome2• 3 years ago

https://huggingface.co/spaces/HuggingFaceH4/open_llm_leaderb... is one place.

amelius2• 3 years ago

How many tokens/second on an average machine?

davidy1233• 3 years ago

Granted I'm not coming from the python world, but I have tried many of these projects, and very few of them install out of the box. They usually end with some incompatibility, and files scattered all over the place, leading to future nightmares.

  ERROR: pip's dependency resolver does not currently take into account all the packages that are installed. This behaviour is the source of the following dependency conflicts.
  sentry-sdk 1.22.2 requires urllib3<2.0.0, but you have urllib3 2.0.2 which is incompatible

Just for fun, here's the result of python -m pip install -r ./requirements.txt for tortoise-tts;

…many many lines

          raise ValueError("%r is not a directory" % (package_path,))
      ValueError: 'build/py3k/scipy' is not a directory
      Converting to Python3 via 2to3...

…

  /tmp/pip-install-hkb_4lh7/scipy_088b20410aca4f0cbcddeac86ac7b7b1/build/py3k/scipy/signal/fir_filter_design.py
      [end of output]

  note: This error originates from a subprocess, and is 
  likely not a problem with pip.
  error: metadata-generation-failed

I'm not asking for support, just saying if people really want to make something 'easy' they'd use docker. I gather there are better python package managers, but I gather that's a bit of a mess too.

Someone is thinking "this is part of learning the language," but I think it's just bad design.

oefrha4• 3 years ago

You don’t need Docker, you just need a virtual env for each random thing you try instead of making them all conflict with each other. Maybe some day pip will add a switch to automatically create one, but until then,

  python3 -m venv venv
  . venv/bin/activate

before you try something random.

Also, `python` is usually Python 2.7. If it is, I advise removing it from your system unless you have a strong reason to keep it.

davidy1233• 3 years ago

Sorry, but

  (base) vid@kk:~/D/ai/tortoise-tts$ python3 -m venv venv
  (base) vid@kk:~/D/ai/tortoise-tts$ . venv/bin/activate
  (venv) (base) vid@kk:~/D/ai/tortoise-tts$ python -m pip install -r ./requirements.txt
  Collecting tqdm
    Using cached tqdm-4.65.0-py3-none-any.whl (77 kB)
  Collecting rotary_embedding_torch
    Using cached rotary_embedding_torch-0.2.3-py3-none-any.whl (4.5 kB)

…

    × python setup.py egg_info did not run successfully.
    │ exit code: 1
    ╰─> [8 lines of output]
        Traceback (most recent call last):
    File "<string>", line 2, in <module>    
    File "<pip-setuptools-caller>", line 34, in <module>    
    File "/tmp/pip-install-i7ubxxkc/scipy_4d5af4f3e2094adca3313ccb41a6d5ff/setup.py", line 196, in <module>    
      setup_package()    
    File "/tmp/pip-install-i7ubxxkc/scipy_4d5af4f3e2094adca3313ccb41a6d5ff/setup.py", line 147, in setup_package    
      from numpy.distutils.core import setup    
        ModuleNotFoundError: No module named 'numpy'
        [end of output]
  
    note: This error originates from a subprocess, and is likely not a problem with pip.
  error: metadata-generation-failed

  × Encountered error while generating package metadata.
  ╰─> See above for output.

  note: This is an issue with the package mentioned above, not pip.
  hint: See above for details.

  [notice] A new release of pip available: 22.3.1 -> 23.1.2
  [notice] To update, run: pip install --upgrade pip
  (venv) (base) vid@kk:~/D/ai/tortoise-tts$

I'm sure you could eventually help get this working, which is kind of you, but the point is the "supposed tos" don't work either. It needs to be comprehensively fixed if python really wants to be approachable. Maybe it doesn't. It's also just not a good citizen when it comes to heterogeneous apps on the same system.

This isn't the first time venv didn't work for me, then there's anaconda, miniconda, and a bunch of other things that add env and directories. I don't really know what any of them do, and -I don't want to- I'm not an expert on every app on my system, but I can use nearly all of them without pain. (remember this is about ease of use)

Oh yeah, and python 2 vs python 3. <rolls eyes>

It's very much the "works for me" experience from the old days. There's no good learning from it, except dependencies suck and python systems aren't good at them.

I think when releasing anything that includes dependencies that span the operating system, it's just good engineering to use a container approach. Otherwise you're just causing a lot of discomfort in the world for no good reason.

It's funny because chatgpt would give me an answer to this in a few moments, but I'm locked out for a week because it can't figure out where I am.

Now I'm spending my Sunday morning setting up a dockerfile for tortoise-tts. At least I will learn something reusable from that. I guess I will create a PR for it, though it seems the author isn't tending the repo anymore.

karlicoss• 3 years ago

Or you can use pipx, it deals with all the virtualenv business behind the scenes

btbuildem2• 3 years ago

Nope, this is EXACTLY why I'd use docker. You want to faff around with some esoteric settings? Go for it! But don't make your would-be users runt the gauntlet, that's pointless.

All that nonsense vs docker pull / docker run

Nephx• 3 years ago

Dangerous comment.

From a linux perspective, I wouldn't blindly suggest the average reader to purge Python 2.7 from their system, as it might drag core parts of the WM with it. Consider aliasing, or better yet, relying on modern venv tools such as Conda instead.

subsaharancoder• 3 years ago

+1 to using venv

gymbeaux• 3 years ago

It’s been a long time since “Python —version” output “2.x” on a computer I was using. Even macOS is on Python 3 these days iirc. Every Linux distro I’ve installed in the last few months was at least 3.8.

anaisbetts• 3 years ago

I would say it's more an artifact of historical tech debt that is hard to change now without breaking everyone. As another commenter pointed out, you want to use a venv - I use pipenv as a tool to automate this but there are others as well (poetry is probably better but pipenv seems to work for me).

seydor5• 3 years ago

This is too much to ask for an OSSLM project , considering that it will be obsolete by something else in 7 days or less.

RcouF1uZ4gsC• 3 years ago

llama.cpp has revolutionized running these LLMs because it provides a nice, self-contained minimal dependency way to do this.

Python is very fragile to deploy and run on your own machine.

ShadowBanThis01• 3 years ago

Yep, I just tried to install a Python-based project and there was a conflict between Pyenv's and Homebrew's versions of pip... despite having used Homebrew to install Pyenv. I ended up just getting rid of Pyenv altogether... but now Python may be in some screwed-up state on my system.

It's too bad the ecosystem seems to be so messy, because Python seems like the best language for general utilities.

replwoacause• 3 years ago

This is exactly why I’ve never wanted to get too involved with Python. The few times I’ve tried to play with it, it always becomes a nightmare of error messages to do with stuff exactly like this and I spent more time farting around trying to figure out what went wrong with it then I do actually doing any development.

tedmiston2• 3 years ago

Yep, I just tried to install a Python-based project and there was a conflict between Pyenv's and Homebrew's versions of pip... despite having used Homebrew to install Pyenv.

The comment does not really make sense. It sounds like pyenv wasn't setup correctly on your system (needs to be added to your bash_profile etc). The typical setup is to put pyenv first in PATH so it takes precedence.

It may be easier to use pipx which will auto manage the virtual envs of end user apps for you.

johnnyjeans• 3 years ago

This is the primary reason I'm averse to languages and ecosystems that rely on package managers. I have never had a good experience where these things aren't just constantly breaking. Stack/cabal, cargo, pip, npm/yarn, gem. Scattering files across my filesystem and having extremely brittle configs that shatter the ecosystem into a billion pieces at seemingly random intervals. A problem exacerbated by these package managers often being more complex than the compiler/interpreter itself. Luarocks is probably the least problematic, and that's mostly because it hosts really simple and self-contained software.

Say what you will about the old school way of manually building and copying shit around, at least when something breaks I don't have to spend a couple hours keelhauling a bloated toolchain in a debugger for mutiny.

tedmiston2• 3 years ago

Are you familiar with virtual environments? It's the standard Python technique for isolating dependencies across projects. [Most projects mention this in the setup / quickstart section of their docs.]

You should not be seeing these dependency conflict issues if you install each project in its own virtual environment.

If you just want them to be easily installed you can just use pipx (`pipx install my-package`) which will manage the virtual environment automatically.

Making a full blown Docker image for it is overkill 99% of the time. Virtual environments serve the same purpose while being much faster and lighter weight.

j_shi2• 3 years ago

Self-hosted + self-trained LLMs are probably the future for enterprise.

While consumers are happy to get their data mined to avoid paying, businesses are the opposite: willing to pay a lot to avoid feeding data to MSFT/GOOG/META.

They may give assurances on data protection (even here GitHub copilot TOS has sketchy language around saving down derived data), but can’t get around fundamental problem that their products need user interactions to work well.

So it seems with BigTechLLM there’s inherent tension between product competitiveness and data privacy, which makes them incompatible with enterprise.

Biz ideas along these lines: - Help enterprises set up, train, maintain own customized LLMs - Security, compliance, monitoring tools - Help AI startups get compliant with enterprise security - Fine tuning service

cddotdotslash• 3 years ago

I suspect the major cloud providers will also each offer their own “enterprise friendly” LLM services (Azure already offers a version of OpenAI’s API). If they have the right data guarantees, that’ll probably be sufficient for companies that are already using their IaaS offerings.

quaintdev• 3 years ago

Enterprises should work on an open source LLM and run it on their own. This also helps people like you and me to run LLM at home.

It has worked before like in case of Linux and can work again.

monkeydust2• 3 years ago

How do the data rights broadly differ between OpenAI API directly and through Azure's endpoint?

SamuelAdams• 3 years ago

In the book “To sleep in a sea of stars” there’s a concept of a “ship mind” that is local to each space craft. It’s smarter than “pseudo ai” and can have real conversations, answer complex questions, and even tell jokes.

I can see a self-hosted LLM being akin to a company’s ship-mind. Anyone can ask questions, order analyses, etc, so long as you are a member of the company. No two LLM’s will be exactly the same - and that’s ok.

https://fractalverse.net/explore-to-sleep-in-a-sea-of-stars/...

londons_explore• 3 years ago

willing to pay a lot to avoid feeding data to MSFT/GOOG/META.

Right now, you can't pay a lot and get a local LLM with similar performance to GPT-4.

Anything you can run on-site isn't really even close in terms of performance.

The ability to finetune to your workplaces terminology and document set is certainly a benefit, but for many usecases that doesn't outweigh the performance difference.

j_shi2• 3 years ago

Open source with own fine tuning closing in fast:

https://lmsys.org/blog/2023-03-30-vicuna/

https://www.semianalysis.com/p/google-we-have-no-moat-and-ne...

superbiome2• 3 years ago

Does something like this exist for local code repos? (Excuse my ignorance since the space is moving faster than light.)

arbol2• 3 years ago

Seems to me that this could be used for exactly that. Just fork the repo and change the filetypes and loaders for your code source files.

arbol2• 3 years ago

https://python.langchain.com/en/latest/modules/indexes/docum...

monkeydust2• 3 years ago

Got this working locally - badly needs GPU support (have a 3090 so come on!) there is some workaround but expect it will come pretty soon. This video was a useful walkthough esp on using different model and upping the CPU threads. https://www.youtube.com/watch?v=A3F5riM5BNE

Havoc• 3 years ago

Would this work better with something like llama or a instruction following model like alpaca?

debbiedowner• 3 years ago

This is a shortcut/workaround to transforming the private docs to a prompt:answer dataset and fine tuning right?

What would be the difference in user experience or information retrieval performance between the two?

My impression is it saves work on the dataset transformation and compute for fine tuning, so it must be less performant. Is there a reason to prefer the strategy here other than ease of setup?

hodanli• 3 years ago

These are the similar projects I've come across:

- [GitHub - e-johnstonn/BriefGPT: Locally hosted tool that connects documents to LLMs for summarization and querying, with a simple GUI.](https://github.com/e-johnstonn/BriefGPT)

- [GitHub - go-skynet/LocalAI: Self-hosted, community-driven, local OpenAI-compatible API. Drop-in replacement for OpenAI running LLMs on consumer-grade hardware. No GPU required. LocalAI is a RESTful API to run ggml compatible models: llama.cpp, alpaca.cpp, gpt4all.cpp, rwkv.cpp, whisper.cpp, vicuna, koala, gpt4all-j, cerebras and many others!](https://github.com/go-skynet/LocalAI)

- [GitHub - paulpierre/RasaGPT: RasaGPT is the first headless LLM chatbot platform built on top of Rasa and Langchain. Built w/ Rasa, FastAPI, Langchain, LlamaIndex, SQLModel, pgvector, ngrok, telegram](https://github.com/paulpierre/RasaGPT)

- [GitHub - imartinez/privateGPT: Interact privately with your documents using the power of GPT, 100% privately, no data leaks](https://github.com/imartinez/privateGPT)

- [GitHub - reworkd/AgentGPT: Assemble, configure, and deploy autonomous AI Agents in your browser.](https://github.com/reworkd/AgentGPT)

- [GitHub - deepset-ai/haystack: Haystack is an open source NLP framework to interact with your data using Transformer models and LLMs (GPT-4, ChatGPT and alike). Haystack offers production-ready tools to quickly build complex question answering, semantic search, text generation applications, and more.](https://github.com/deepset-ai/haystack)

- [PocketLLM « ThirdAi](https://www.thirdai.com/pocketllm/)

- [GitHub - imClumsyPanda/langchain-ChatGLM: langchain-ChatGLM, local knowledge based ChatGLM with langchain ｜基于本地知识库的 ChatGLM 问答](https://github.com/imClumsyPanda/langchain-ChatGLM)

zestyping• 3 years ago

Would someone do me the kindness of explaining (a little more) how this works?

It looks like you can ask a question and the model will use its combined knowledge of all your documents to figure out the answer. It looks like it isn't fine-tuned or trained on all the documents, is that right? How is each document turned into an embedding, and then how does the model figure out which documents to consult to answer the question?

PrivateGPT

Related Stories