jnjj/gemma-3-1b-it-qat-int4-quantized-inference-unrestricted-weights-only-sf

Upload INT4 quantized Gemma‑3‑1B‑IT QAT with bfloat16 compute, extensive unconventional modifications including instruction conversion flag, and only bfloat16 .weight tensors saved as safetensors (bfloat16 compute)

Create README.md

Upload INT4 quantized Gemma‑3‑1B‑IT QAT with bfloat16 compute, extensive unconventional modifications, and only bfloat16 .weight tensors saved as safetensors (bfloat16 compute)

initial commit