pith:SCKZ2TKU
PermuQuant: Lowering Per-Group Quantization Error by Reordering Channels for Diffusion Models
Reordering channels to group similar statistics reduces per-group quantization error in diffusion models.
arxiv:2605.09503 v2 · 2026-05-10 · cs.CV
Add to your LaTeX paper
\usepackage{pith}
\pithnumber{SCKZ2TKU7R5RXVZCWPLVTA645T}
Prints a linked badge after your title and injects PDF metadata. Compiles on arXiv. Learn more · Embed verified badge
Record completeness
Claims
PermuQuant consistently reduces quantization error and outperforms existing PTQ baselines. On FLUX.1-dev with an RTX 5090, PermuQuant achieves up to a 1.8× single step speedup and reduces the DiT memory footprint by 3.5× under W4A4 NVFP4 quantization.
That a permutation chosen on calibration data via the joint second-moment criterion will generalize to the full input distribution at inference time and will not introduce new artifacts in the generated outputs.
PermuQuant reduces per-group quantization error in diffusion models by sorting channels with similar activation and weight statistics into the same groups using a calibration-checked permutation.
Cited by
Receipt and verification
| First computed | 2026-06-02T02:04:18.799441Z |
|---|---|
| Builder | pith-number-builder-2026-05-17-v1 |
| Signature | Pith Ed25519
(pith-v1-2026-05) · public key |
| Schema | pith-number/v1.0 |
Canonical hash
90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6
Aliases
· · · · ·Agent API
Verify this Pith Number yourself
curl -sH 'Accept: application/ld+json' https://pith.science/pith/SCKZ2TKU7R5RXVZCWPLVTA645T \
| jq -c '.canonical_record' \
| python3 -c "import sys,json,hashlib; b=json.dumps(json.loads(sys.stdin.read()), sort_keys=True, separators=(',',':'), ensure_ascii=False).encode(); print(hashlib.sha256(b).hexdigest())"
# expect: 90959d4d54fc7b1bd722b3d75983dcecc550c732931eedeef759a807c0dbb2f6
Canonical record JSON
{
"metadata": {
"abstract_canon_sha256": "89dd5604c2002a67a55cb5083eb65736dcd9dbddfb035fcb21d6ef526dbfe745",
"cross_cats_sorted": [],
"license": "http://creativecommons.org/licenses/by/4.0/",
"primary_cat": "cs.CV",
"submitted_at": "2026-05-10T12:26:50Z",
"title_canon_sha256": "cfdfec046d9e408ceb2326dacfe0e3d66d76b14d219b49053b4cf07042f68767"
},
"schema_version": "1.0",
"source": {
"id": "2605.09503",
"kind": "arxiv",
"version": 2
}
}