RahulSChand/gpu_poor
▲ +3 ★ this week
Calculate token/s & GPU memory requirement for any LLM. Supports llama.cpp/ggml/bnb/QLoRA quantization
LLaMA 2 implemented from scratch in PyTorch
$ git clone https://github.com/hkproj/pytorch-llama.git▲ +3 ★ this week
Calculate token/s & GPU memory requirement for any LLM. Supports llama.cpp/ggml/bnb/QLoRA quantization
▲ +1 ★ this week
On-device LLM Inference Powered by X-Bit Quantization
Code for ACL 2024 paper "TruthX: Alleviating Hallucinations by Editing Large Language Models in Truthful Space"
Text classification with Foundation Language Model LLaMA
▲ +5 ★ this week
RWKV (pronounced RwaKuv) is an RNN with great LLM performance, which can also be directly trained like a GPT transformer (parallelizable). We are at RWKV-7 "Goose". So it's combining the best of RNN and transformer - great performance, linear time, constant space (no kv-cache), fast training, infinite ctx_len, and free sentence embedding.
Data from GitHub · snapshot Nov 16, 2026