Megatron-LM VS TensorRT

# Install necessary dependencies sudo apt update sudo apt install python3-pip # Install PyTorch with GPU support pip3 install torch torchvision torchaudio --extra-index-url https://download.pytorch.org/whl/cu113 # Clone Megatron LM repository git clone https://github.com/NVIDIA/Megatron-LM.git cd Megatron-LM # Install Megatron LM dependencies pip3 install -r requirements.txt # Install NVIDIA Apex for mixed-precision training git clone https://github.com/NVIDIA/apex cd apex pip3 install -v --disable-pip-version-check --no-cache-dir ./

https://github.com/NVIDIA/Megatron-LM

This is probably a good baseline to start thinking about LLM training at scale.

This 20B model was trained on the same datasets as its predecessor, aptly named The Pile. Furthermore, the libraries Megatron and DeepSpeed were used to achieve better computing resource utilization, and eventually GPT-NeoX evolved into its own framework for training other LLMs. It was used, for example, as the foundation for Llemma, an open-source model specializing on theorem proving.

Deepspeed Megatron-LM

GPU cluster scaling has come a long way. Just checkout the scaling plot here: https://github.com/NVIDIA/Megatron-LM

I found their code that the softmax function conduct all-reduce before they work.

Google's PaLM, Facebook's LLaMA, Nvidia's Megatron, I am missing some surely and Apple sure has something cooking as well but these are the big ones, of course none of them are publicly available, but research papers are reputable. All of the ones mentioned should beat GPT-3 although GPT-3.5 (chatGPT) should be bit better and ability to search (Bing) should level the playing field even further, but Google's PaLM with search functionality should be clearly ahead. This is why people are excited about GPT-4, GPT-3 was way ahead of anyone else when it came out but others were able to catch up since, we'll see if GPT-4 will be another bing jump among LLMs.

What is the purpose of using ONNX or TensorRT for deployment? When deploying a trained deep learning model into a real-world service environment for inference, optimization to increase execution speed and reduce resource consumption is crucial. ONNX and TensorRT are prominent tools and frameworks widely used for this purpose.

Extensions: Jan supports extensions like TensortRT and Inference Nitro for customizing and enhancing your AI models.

(python demo_txt2img.py "a beautiful photograph of Mt. Fuji during cherry blossom"). https://github.com/NVIDIA/TensorRT/tree/release/8.6/demo/Diffusion

- https://github.com/NVIDIA/TensorRT

TVM and other compiler-based approaches seem to really perform really well and make supporting different backends really easy. A good friend who's been in this space for a while told me llama.cpp is sort of a "hand crafted" version of what these compilers could output, which I think speaks to the craftmanship Georgi and the ggml team have put into llama.cpp, but also the opportunity to "compile" versions of llama.cpp for other model architectures or platforms.

https://github.com/NVIDIA/TensorRT/issues/982

Maybe? Looks like tensorRT does work, but I couldn't find much.

highly optimized transformer-based encoder and decoder component, supported on pytorch, tensorflow and triton

TensorRT, custom ml framework/ inference runtime from nvidia, https://developer.nvidia.com/tensorrt, but you have to port your models

It uses the nvidia demo code from: https://github.com/NVIDIA/TensorRT/tree/main/demo/Diffusion