Model Compression, Quantization & On-Device ML Quiz

Test your understanding of scale and zero-point, per-channel versus per-tensor quantization, calibration, PTQ versus QAT and the straight-through estimator, LLM activation outliers, GPTQ, AWQ and SmoothQuant, memory-bound decode arithmetic, structured versus unstructured pruning, distillation temperature, LoRA merging, ONNX export pitfalls, benchmarking and the on-device decision.

Start quiz