Reza2kn/Audio8-ASR-Infinite-Compressed

Verify fresh-download MLX memory runtime in English and Chinese

Reduce MLX cache-rebase allocation and report sustained GPU streaming

Verify downloaded batch2 Mac streaming launcher

Keep evaluation inputs external to the model release

Lower Mac streaming batch size after separate paired qualification

Verify fresh public CPU, MLX, CUDA and Vulkan consumers

Release complete mixed Q3/Q4 Audio8 with portable CPU, MLX and exact GGUF GPU consumers

Publish complete experimental Audio8 Q4 model and runtimes

Publish complete experimental Audio8 Q4 model and runtimes

initial commit