Back
A report + reflection on building demo AI quoting system by fine-tuning Qwen3-4B with a single RTX 4090, quantized to Q4_K_M GGUF, and benchmarked CPU vs GPU.
llm
fine-tuning
quantization
rtx 4090
