Skip to content

Latest commit

 

History

History
148 lines (108 loc) · 5.2 KB

README.md

File metadata and controls

148 lines (108 loc) · 5.2 KB

WasmGPT

ChatGPT-like chatbot in the browser using ggml and emscripten. No API keys required. No server required. No data is sent to any server.

This demo uses a Cerebras-GPT-1.3B-Alpaca-SP, which is a version of the Cerebras-GPT-1.3B model LoRA-finetuned on the Alpaca dataset.

Limitations

  • The model is very hallucinatory and can generate very incorrect text.
  • I limited initial system prompt and removed memory/context for speed.
  • The model is around 900MB and takes a while to load.
  • Doesn't work on mobile safari. Probably won't work in all browsers.
  • ggml's implementation of gpt2 doesn't have repetition penalty (TODO), making it very repetitive

TODOs

  • Less unhinged model support
  • Repetition penalty
  • Longer context

How to run locally

  1. Have emscripten installed and activated

  2. Clone this repo:

git clone -b wasm-demo https://github.com/lxe/ggml.git
cd ggml
  1. In order to make wasm work, you need to serve it over https and provide Cross-Origin-Embedder-Policy and Cross-Origin-Opener-Policy headers. You can use the following commands to generate a self-signed certificate.
openssl req -x509 -newkey rsa:4096 -keyout key.pem -out cert.pem -days 365 -nodes \
-subj "/CN=localhost" -addext "subjectAltName = DNS:localhost"

You will also need to add the certificate to your browser's root store.

  1. Build the software and run the server:
mkdir build
cd build
emcmake cmake ..
make gpt-2
cd .. && python server.py

ggml

Tensor library for machine learning

Note that this project is under development and not ready for production use.
Some of the development is currently happening in the llama.cpp and whisper.cpp repos

Features

  • Written in C
  • 16-bit float support
  • 4-bit integer quantization support
  • Automatic differentiation (WIP in progress)
  • ADAM and L-BFGS optimizers
  • Optimized for Apple silicon via NEON intrinsics and Accelerate framework
  • On x86 architectures utilzes AVX intrinsics
  • No third-party dependencies
  • Zero memory allocations during runtime

Roadmap

Whisper inference (example)

With ggml you can efficiently run Whisper inference on the CPU.

Memory requirements:

Model Disk Mem
tiny 75 MB ~280 MB
base 142 MB ~430 MB
small 466 MB ~1.0 GB
medium 1.5 GB ~2.6 GB
large 2.9 GB ~4.7 GB

GPT inference (example)

With ggml you can efficiently run GPT-2 and GPT-J inference on the CPU.

Here is how to run the example programs:

# Build ggml + examples
git clone https://github.com/ggerganov/ggml
cd ggml
mkdir build && cd build
cmake ..
make -j4 gpt-2 gpt-j

# Run the GPT-2 small 117M model
../examples/gpt-2/download-ggml-model.sh 117M
./bin/gpt-2 -m models/gpt-2-117M/ggml-model.bin -p "This is an example"

# Run the GPT-J 6B model (requires 12GB disk space and 16GB CPU RAM)
../examples/gpt-j/download-ggml-model.sh 6B
./bin/gpt-j -m models/gpt-j-6B/ggml-model.bin -p "This is an example"

# Run the Cerebras-GPT 111M model
# Download from: https://huggingface.co/cerebras
python3 ../examples/gpt-2/convert-cerebras-to-ggml.py /path/to/Cerebras-GPT-111M/
./bin/gpt-2 -m /path/to/Cerebras-GPT-111M/ggml-model-f16.bin -p "This is an example"

The inference speeds that I get for the different models on my 32GB MacBook M1 Pro are as follows:

Model Size Time / Token
GPT-2 117M 5 ms
GPT-2 345M 12 ms
GPT-2 774M 23 ms
GPT-2 1558M 42 ms
--- --- ---
GPT-J 6B 125 ms

For more information, checkout the corresponding programs in the examples folder.

Using cuBLAS

# fix the path to point to your CUDA compiler
cmake -DGGML_CUBLAS=ON -DCMAKE_CUDA_COMPILER=/usr/local/cuda-12.1/bin/nvcc ..