This repository contains GGUF format model files forQwen’s Qwen3.6-27B. Qwen3.6-27B is a highly capable 27B parameter model featuring a Hybrid Gated DeltaNet + Gated Attention architecture and a unified vision-language foundation. It natively supports up to 262,144 tokens of context, tool calling, multimodal inputs (images/video), and introduces «Thinking Preservation» to retain reasoning context across historical messages for improved agentic coding workflows. To see the available files, please verify the Files and versions tab. Important: This model uses a hybrid Gated DeltaNet architecture. Ensure you are using the absolute latest version of llama.cpp to support these new operators. Recommended Parameters: Qwen3.6 operates in Thinking Mode by default. Temperature: 1.0 (Thinking/Reasoning) or 0.6 (Precise coding tasks) Top-P: 0.95 Context: -c` (The model supports ultra-long context natively, but 32k-128k is recommended for most local use to conserve VRAM/RAM). > [!Note] > This repository contains model weights and configuration files for the post-trained model in the Hugging Face Transformers format. > > These artifacts are compatible with Hugging Face Transformers,…
Модальности:
Генерация текста Компьютерное зрение Мультимодальность
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: AaryanK
Теги: gguf, text-generation-inference, qwen, multimodal, vision, agent, en, endpoints_compatible
Лайков: 3 | Загрузок: 531
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
