chaoliangUNSW/Jev-Style-Qwen3.5-2B-Decision-v2-GGUF

new_version -> chaoliangUNSW/Jev-Style-2B-Decision-v3-GGUF (same 2B size class)

Card metadata: new_version points to the v3 0.8B release

README: add Ollama usage; SHA256SUMS: update README, add template/params

Add Ollama template override (auto-converted template dropped the prompt)

Map current GGUF filenames to validated build names in quantization_summary.json

Rename GGUF files so the quant type ends the filename (Ollama :Q4_K_M tags)

Add website link (jevstyle.com) to model card

Fix contact email typo in model card (yanchaoliang369@gmail.com)

Add calibrated Q4_K_M and BF16 GGUF download options

Add benchmark, calibration and robustness charts with result-first model cards

Add benchmark, calibration and robustness charts with result-first model cards

Release Jev-Style v2 with calibrated decision inference and evaluation records

initial commit