Qwen发布Qwen3.8-Flash-Next-FP8开放权重
Qwen在Hugging Face上发布了Qwen3.8-Flash-Next的FP8量化权重和配置文件，采用Transformers格式。Qwen称其采用块大小为128的细粒度FP8量化，性能指标与原模型接近。仓库还列出对Transformers、vLLM、SGLang和TokenSpeed的兼容性，但所给报告没有独立的兼容性测试或基准结果。
此次发布为开发者提供了Qwen3.8-Flash-Next的FP8开放权重文件和配置说明。所给报告尚未独立验证Qwen关于性能和框架兼容性的说法。
Qwen/Qwen3.8-Flash-Next-FP8 Image-Text-to-Text • 180B • Updated 3 days ago • 142k • • 304
base_model:
- Qwen/Qwen3.8-Flash-Next library_name: transformers license: other license_name: qwen-community-1.0 license_link: LICENSE pipeline_tag: image-text-to-text
Qwen3.8-Flash-Next-FP8
[!Note] This repository contains FP8-quantized model weights and configuration files for the post-trained model in the Hugging Face Transformers format. These artifacts are compatible with Hugging Face Transformers, vLLM, SGLang, TokenSpeed, etc. The quantization method is fine-grained fp8 quantization with block size of 128, and its performance metrics are nearly identical to those of the original model.
#Qwen#Qwen3.8-Flash-Next-FP8#Qwen3.8-Flash-Next#Hugging Face Transformers#vLLM#SGLang#TokenSpeed#Qwen Cloud