mewse/gemma-4-26B-A4B-it-qat-q4_0-unquantized-heretic-ara
2596
Delete gemma-4-26B-A4B-heretic-ara-en.i1-IQ4_XS.gguf
add i1-IQ4_XS GGUF (imatrix-quantized, ready for llama.cpp)
remove xet rate test
xet rate test (will delete)
remove rate test
rate test (will delete)
remove rate test
rate test (will delete)
remove adapter-only files (full merged weights in repo)
remove adapter-only files (full merged weights in repo)
cleanup
add refusal eval table vs other builds
remove rate test
rate test (will delete)
add refusal eval table vs other builds
add full merged bf16 weights (ARA trial #60, alpha/r=1.0)
Upload README.md with huggingface_hub
Upload tokenizer
Upload model
initial commit