DeepSeek-R1 7B INT4 at 69.3 tok/s on a $300 RTX 3060. Faster than llama.cpp, vLLM, and NVIDIA TensorRT-LLM. Is one developer + Ai really better than the entire industry?

GLM-5.2 744B at 4-bit on Modal 4x H200 via vLLM, plus a static streaming chat UI.

Exact minimax FP16 scale selection for multi-environment W4A16 quantization, with reproducible Qwen3.5 evidence.