An MLX port of tencent/Hy3 (Tencent Hunyuan 3, hyv3`), a 295B-parameter Mixture-of-Experts model (21B active, 192 experts top-8, GQA + QK-norm, 256K context) with a Multi-Token-Prediction (MTP) layer for self-speculative decoding. Quantized in the asymmetric style of antirez/deepseek-v4-gguf: crush the routed experts (they are most of the weights), keep the decision-making parts high precision. The filename is the spec. Runs on a single 128 GB Apple Silicon machine. INVALID LANGUAGE PAIR SPECIFIED. EXAMPLE: LANGPAIR=EN|IT USING 2 LETTER ISO OR RFC3066 LIKE ZH-CN. ALMOST ALL LANGUAGES SUPPORTED BUT SOME MAY HAVE NO CONTENT The transformers pin avoids a known crash on 5.13+ (AutoTokenizer.register`). > Read this before running (memory). The weights must stay fully resident in RAM. > This model is ~110 GB on a 128 GB Mac, which is tight. For usable speed: > 1. Keep the model on the internal SSD (external USB works but…
Модальности:
Генерация текста
Области применения:
Диалог / чат Следование инструкциям
Задача: Генерация текста
Автор: ox-ox
Теги: mlx, hy_v3, hunyuan, hy3, moe, mixture-of-experts, quantized, 2-bit
Лайков: 3 | Загрузок: 474
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
