VnimanieAI/Qwen3.8-Flash-Next-W4A16

Recommend cyankiwi AWQ for 8x24GB setups; add head-to-head results; attribute verbosity to quant recipe

Correct GPQA budget-corrected accounting (0.854 -> 0.874); clarify hosted reference

Add capability benchmarks (tau2-bench Airline, MMLU_RU, GPQA Diamond)

library_name transformers: enable Hub vLLM app-compatibility (transformersInfo)

Add 4-bit tag for Hub precision filter

Upload folder using huggingface_hub

Upload LICENSE with huggingface_hub

Upload README.md with huggingface_hub

Add files using upload-large-folder tool

initial commit