Popular repositories Loading

1. dflash dflash Public

   DFlash: Block Diffusion for Flash Speculative Decoding

   Python 5.2k 374

2. paroquant paroquant Public

   [ICLR 2026] ParoQuant: Pairwise Rotation Quantization for Efficient Reasoning LLM Inference

   Python 309 30

3. sparselora sparselora Public

   [ICML 2025] SparseLoRA: Accelerating LLM Fine-Tuning with Contextual Sparsity

   Python 76 6

4. flash-colreduce flash-colreduce Public

   Fast, memory-efficient attention column reduction (e.g., sum, mean, max)

   Python 48 2