mirror of
https://github.com/nomic-ai/gpt4all
synced 2024-11-02 09:40:42 +00:00
67fcfeea8b
Users may want to play around with gpt4all-j from the command line. But they may not have Qt, and might not want to get it, or may find it very hard to do so - eg when using a Google Colab or similar hosted service. It's easy to build the CLI tools just by building the `ggml` sub folder. So this commit adds instructions on doing that, including an example invocation of the `gpt-j` binary. |
||
|---|---|---|
| cmake | ||
| ggml@c9f702ac57 | ||
| icons | ||
| .gitignore | ||
| .gitmodules | ||
| CMakeLists.txt | ||
| gptj.cpp | ||
| gptj.h | ||
| LICENSE | ||
| llm.cpp | ||
| llm.h | ||
| llmodel.h | ||
| main.cpp | ||
| main.qml | ||
| README.md | ||
gpt4all-chat
Cross platform Qt based GUI for GPT4All versions with GPT-J as the base model. NOTE: The model seen in the screenshot is actually a preview of a new training run for GPT4All based on GPT-J. The GPT4All project is busy at work getting ready to release this model including installers for all three major OS's. In the meantime, you can try this UI out with the original GPT-J model by following build instructions below.
Features
- Cross-platform (Linux, Windows, MacOSX)
- Fast CPU based inference using ggml for GPT-J based models
- The UI is made to look and feel like you've come to expect from a chatty gpt
- Check for updates so you can alway stay fresh with latest models
- Easy to install with precompiled binaries available for all three major desktop platforms
Feature wishlist
- Multi-chat - a list of current and past chats and the ability to save/delete/export and switch between
- Text to speech - have the AI response with voice
- Speech to text - give the prompt with your voice
- Multi-modal - Ability to load more than one model and switch between them
- Python bindings
- Typescript bindings
- Plugin support for langchain other developer tools
- Save your prompt/responses to disk
- Upload prompt/respones manually/automatically to nomic.ai to aid future training runs
- Syntax highlighting support for programming languages, etc.
- REST API with a built-in webserver in the chat gui itself with a headless operation mode as well
- Advanced settings for changing temperature, topk, etc. (DONE)
- YOUR IDEA HERE
Building and running
- Install Qt 6.x for your platform https://doc.qt.io/qt-6/get-and-install-qt.html
- Install cmake for your platform https://cmake.org/install/
- Download https://huggingface.co/EleutherAI/gpt-j-6b
- Clone this repo and build
git clone --recurse-submodules https://github.com/nomic-ai/gpt4all
cd gpt4all-chat
mkdir build
cd build
cmake ..
cmake --build . --parallel
python3 ../ggml/examples/gpt-j/convert-h5-to-ggml.py /path/to/your/local/copy/of/EleutherAI/gpt-j-6B 0
./bin/gpt-j-quantize /path/to/your/local/copy/of/EleutherAI/gpt-j-6B/ggml-model-f32.bin ./ggml-model-q4_0.bin 2
./chat
Building and running CLI tools only (no Qt required)
- Install cmake for your platform https://cmake.org/install/
- Clone this repo and build the
ggmlsubfolder
git clone --recurse-submodules https://github.com/nomic-ai/gpt4all
cd gpt4all-chat/ggml
mkdir build
cd build
cmake ..
cmake --build . --parallel
wget https://gpt4all.io/models/ggml-gpt4all-j.bin # Download GGML model if required
./bin/gpt-j -m ggml-gpt4all-j.bin -n 200 --top_k 40 --top_p 0.9 -b 9 --temp 0.9 -p "Below is an instruction that describes a task. Write a response that appropriately completes the request.
### Instruction:
Write a story about llamas
### Response:"
Contributing
- Pull requests welcome. See the feature wish list for ideas :)
License
The source code of this chat interface is currently under a GPL license. The underlying GPT4All-j model is released under non-restrictive open-source Apache 2 License.
The GPT4All-J license allows for users to use generated outputs as they see fit. Users take responsibility for ensuring their content meets applicable requirements for publication in a given context or region.