
MODELS
by Qwen
Qwen3.6-27B-FP8 is a 27.78B-parameter Qwen image-text-to-text model variant distributed with FP8 artifacts.
Qwen3.6-27B-FP8 is listed by Qwen on Hugging Face as an image-text-to-text model and by Qwen on ModelScope as a 27.78B-parameter FP8 model with Safetensors/PyTorch artifacts. The official QwenLM/Qwen3.6 GitHub repository describes the Qwen3.6 model family and notes that Qwen3.6-27B became available on Hugging Face Hub and ModelScope on 2026-04-22, with quickstart, deployment, and licensing notes. ModelScope includes serving commands for the exact Qwen/Qwen3.6-27B-FP8 model ID.
Use when you need to evaluate a Qwen3.6 27B-class image-text-to-text model distributed in FP8 form. Use when you want model artifacts available through Hugging Face Hub or ModelScope for testing, inspection, or deployment workflows. Use when your serving workflow can follow the ModelScope-provided vLLM commands for the Qwen/Qwen3.6-27B-FP8 model ID.
Last updated Jun 5, 2026