LIVE
Hugging FacePublisher: qwen

Qwen3.8-27B-FP8

Model overview

What this model is for

Qwen3.8-27B-FP8 is a 27.8-billion-parameter Hugging Face repository for image-text-to-text processing, making it relevant to developers building applications that accept images and text and return text. Its listed modalities are text and image, and its supplied use case is vision. The repository identifies the model as an FP8 publication and provides access to its files and execution options through Hugging Face. This makes it a candidate for image understanding, visual question answering, image description, and other multimodal text-generation workflows, although those specific application patterns are practical interpretations of its advertised image-text-to-text capability rather than documented benchmark results. The supplied material does not provide an advertised context window, inference speed, pricing, benchmark scores, detailed capability tags, or runtime requirements, so these should be verified before deployment. The model is described as available under the Apache-2.0 license; developers should still review the repository’s complete license and usage terms for applicable conditions. The exact parameter count is 27,781,427,952, while the source’s rounded description says 28B parameters.

Classification

Categories, modalities and capabilities

Use-case categories

Input and output modalities

textimage

Supported capabilities

Not disclosed

Continue exploring