Skip to content

Qwen3.8-Flash-Next

Quantization examples for Qwen3.8's Flash sparse MoE model.

Note: Qwen3.8-Flash-Next requires transformers >= 5.17.0, which can be installed with:

uv pip install --upgrade "transformers>=5.17.0"

Examples

Pre-quantized Checkpoints

Model Format Hugging Face Link
Qwen3.8-Flash-Next NVFP4 RedHatAI/Qwen3.8-Flash-Next-NVFP4