Compressed checkpoint: 3,450,632,192 bytes = 0.3564 of the bf16 reference (9,682,900,992 bytes), measured with count_total_size (numel * itemsize over the state dict), which eval_submission.py documents as the graded quantity.

Model: Qwen/Qwen3.5-4B. Domain: math.

Per evaluation set:

Measured on our own frozen benchmark (four math sets, 16384 new tokens, greedy), not on the hidden set.

artifact.pt (3.45 GB) is in the HuggingFace repo, not here — GitHub caps files at 100 MB. code.py loads it from its own directory.

This recipe is post-training only, so code.py can rebuild the checkpoint from the base model. It ships the measured one anyway: GPTQ is not bit-reproducible across runs — two builds of this recipe from identical source differ in 602 of 834 stored tensors, because the Hessian is measured on freshly drawn calibration text and GPU reductions do not re-associate the same way twice. Shipping the artifact means the graded checkpoint is exactly the one the numbers above describe, and saves ~200 GPU-minutes of recompression.

Calibration: 128 sequences from the eval-shaped mixture (50% competition maths, 25% grade-school, 25% multiple choice), matching the format split of the four graded sets rather than free-form derivations alone.

A random Hadamard transform is applied per weight block before quantisation and inverted at restore, so GPTQ's error compensation runs in an incoherent basis where outliers are spread across channels instead of concentrated in a few.

Track 2 relevance: every stage is a uniform grouped-affine integer code with a fixed group size. Dequantisation is a scale-and-shift over contiguous groups and the Hadamard transform is a butterfly — both are standard fused-dequant GEMM kernel shapes, so nothing here relies on data-dependent lookup or irregular sparsity that would block a CUDA speedup.

artifact.pt must sit next to code.py.