Popular repositories Loading

1. Adaptive-Medusa Adaptive-Medusa Public

   Forked from FasterDecoding/Medusa

   Adaptive-Medusa: Improving Speculative Decoding Performance by Pruning Candidates

   Jupyter Notebook

2. quack quack Public

   Forked from Dao-AILab/quack

   A Quirky Assortment of CuTe Kernels

   Python

3. lm-engine lm-engine Public

   Forked from open-lm-engine/lm-engine

   LM engine is a library for pretraining/finetuning LLMs

   Python

4. flash-attention flash-attention Public

   Forked from Dao-AILab/flash-attention

   Fast and memory-efficient exact attention

   Python

5. Qwen3 Qwen3 Public

   Forked from QwenLM/Qwen3

   Qwen3 is the large language model series developed by Qwen team, Alibaba Cloud.

   Python

6. dion dion Public

   Dion optimizer algorithm

   Python