Repository navigation

ggml-hrx: MXFP4 weights (shared dequantizer, exact E8M0 scale, known-answer test) - #65

Merged

bong-water-water-bong merged 5 commits into

1bit/hrx-vulkan-patchedfrom

1bit/hrx-mxfp4-v2

Oct 2, 2026

Commits on Oct 2, 2026
- ggml-hrx: MXFP4 weights in the shared dequantizer
  
  bong-water-water-bong
  and
  claude
  committed
  89ddd94
- ggml-hrx: exact E8M0 half scale for MXFP4 in the shared dequantizer
  
  bong-water-water-bong
  and
  claude
  committed
  6892003
- tests: test-hrx-mxfp4, a bit-exact known-answer check of MXFP4 GET_ROWS on HRX
  
  bong-water-water-bong
  and
  claude
  committed
  b0d3d46
- tests: test-hrx-mxfp4 accepts zeros only for rows whose E8M0 scale is subnormal
  
  bong-water-water-bong
  and
  claude
  committed
  52b79b9
- ggml-hrx: MUL_MAT_ID admits only input sizes that are a multiple of 256
  
  bong-water-water-bong
  and
  claude
  committed
  adfd541