hanlinxuy's repositories
RWKV-LM
RWKV is an RNN with transformer-level LLM performance. It can be directly trained like a GPT (parallelizable). So it's combining the best of RNN and transformer - great performance, fast inference, saves VRAM, fast training, "infinite" ctx_len, and free sentence embedding.
RWKV_UNKNOWN
RWKV implementation with pure torch and deepspeed.
Awesome-LLM-Inference
📖A curated list of Awesome LLM Inference Paper with codes, TensorRT-LLM, vLLM, streaming-llm, AWQ, SmoothQuant, WINT8/4, Continuous Batching, FlashAttention, PagedAttention etc.
HASCO
agile hardware-software co-design
nn_vis
A project for processing neural networks and rendering to gain insights on the architecture and parameters of a model through a decluttered representation.
numpy
The fundamental package for scientific computing with Python.
Pheno_HTM
The madgraph controller with multicore processing by python. Also include some analysis interface with using matplotlib
PriconneScrapper
Princess Connect ReDive info bot for discord.
pytorch
Tensors and Dynamic neural networks in Python with strong GPU acceleration
QuIP_for_rwkv
Code for paper: "QuIP: 2-Bit Quantization of Large Language Models With Guarantees"
RPTQ4LLM
Reorder-based post-training quantization for large language model
RWKV-Ouroboros
This project is established for real-time training of the RWKV model.
torchscale
Foundation Architecture for (M)LLMs