Shinrali/Qwen3.8-Flash-Next-NVIDIA-Hybrid-Single-Spark

Separate BF16 reference model card

Document RTX PRO 6000 WSL2 benchmark

Clarify BF16 reference and project navigation

Fix multimodal pipeline metadata

Link faster FP8 LM head release and add acceptance estimates

Report third local quality run

Report repeated local quality regression

Focus model card on this NVIDIA hybrid release

Document downloadable converted checkpoint

Upload converted NVIDIA hybrid checkpoint

Publish NVIDIA hybrid single-Spark recipe and model card

initial commit