> [!IMPORTANT] > Naming notice (2026-04-10). The «HLWQ» technique used in this model is being rebranded to HLWQ (Hadamard-Lloyd Weight Quantization). The change is only the name; the algorithm and the weights in this repository are unchanged. > > The rebrand resolves a name collision with an unrelated, earlier KV cache quantization method also named HLWQ (Han et al., arXiv:2502.02617, 2025). HLWQ addresses weight quantization with a deterministic Walsh-Hadamard rotation and Lloyd-Max scalar codebook; Han et al.’s HLWQ addresses KV cache quantization with a random polar rotation. The two methods are technically distinct. > > Existing loaders that load this repository by ID continue to work without changes. Future model uploads will use the HLWQ name. > > Reference paper for this technique: arXiv:2603.29078 (v2 in preparation; v1 still uses the old name). HLWQ Q5 preprocessing produces better INT4 weights than direct quantization — stored in CompressedTensors format for native vLLM inference. That’s it. No plugin, no pip install polarquant, no custom code. Standard INT4 (GPTQ/AWQ) quantizes weights directly — outliers cause errors. 1. Hadamard rotation — distributes weight energy…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: caiovicentino1
Теги: qwen3_5, image-text-to-text, hlwq, quantized, compressed-tensors, int4, marlin, vllm
Лайков: 3 | Загрузок: 36
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
