models qwen3.8 27b cuda gpu - Azure/azureml-assets GitHub Wiki

qwen3.8-27b-cuda-gpu

Overview

This model is an optimized version of Qwen3.8-27B to enable local text inference on CUDA GPUs. This model uses RTN quantization with DFlash2 speculative decoding and requires Foundry Local 2.2.1 or later.

Model Description

  • Developed by: Microsoft
  • Model type: ONNX
  • License: apache-2.0
  • Model Description: This is a conversion of the Qwen3.8-27B for local text inference on CUDA GPUs.
  • Disclaimer: Model is only an optimization of the base model, any risk associated with the model is the responsibility of the user of the model. Please verify and test for your scenarios. There may be a slight difference in output from the base model with the optimizations applied. Note that optimizations applied are distinct from fine tuning and thus do not alter the intended uses or capabilities of the model.

Base Model Information

See Hugging Face model Qwen3.8-27B for details.

Version: 1

Tags

capabilities : reasoning,tool-calling supportsReasoning : true contextLength : 262144

View in Studio: https://ml.azure.com/registries/azureml/models/qwen3.8-27b-cuda-gpu/version/1

⚠️ **GitHub.com Fallback** ⚠️