HybridDiffusion-9B is an 8.95B-parameter text-generation checkpoint produced by applying HybridDiffusion’s autoregressive-to-diffusion conversion recipe to Qwen3.5-9B. A single set of weights supports AR-Trust verified decoding, Diffusion-Trust parallel denoising, and ordinary causal generation through the HybridDiffusion runtime. The native mask ID is 248077. It belongs to the padded model vocabulary and is intentionally not exposed as a tokenizer token. HybridDiffusion inference uses the SGLang and FlashInfer implementations bundled with the HybridDiffusion repository. The checkpoint uses a standard Hugging Face layout and is downloaded automatically by the serving runtime. The one-shot installer creates the evaluation environment and places reusable Hugging Face, FlashInfer, Triton, and extension caches below HYBRIDDIFFUSIONCACHEROOT`. Start one decoding mode. Each command below serves the same checkpoint and uses port 30000 unless PORT is overridden. To run more than one server, assign a different PORT and GPU to each process, for example PORT=30001 CUDAVISIBLEDEVICES=1 …. The installed environment also includes the OpenAI Python client: AR-Trust uses speculation horizon N=4…
Модальности:
Генерация текста
Области применения:
Диалог / чат
Задача: Генерация текста
Автор: yuchen-zhu-zyc
Теги: sglang, qwen3_5, hybrid-diffusion, diffusion-language-model, hybrid-attention, self-speculative-decoding, qwen3.5, conversational
Лайков: 4 | Загрузок: 645
Описание основано на материалах HuggingFace. Перевод выполнен автоматически.
