Qwen 3.8 flash next
Running the Qwen3.8-Flash-Next model
On discrete GPU with a large system RAM
- Basalt : Local inference engine for Qwen3.8-Flash-Next on NVIDIA Blackwell.
- PolyStrata : supports various MoE models.
- Strata : runs Qwen3.8-Flash-Next and DeepSeekV4.1-Flash-1017 from 12GB+ VRAM with smart expert caching.
On one DGX spark
Others
- colab-qwen38-g4 : runs Q38FN on a Google Colab G4 GPU (96 GB VRAM).
Last updated on