Revision History
GPTQ · 6 revisions
Sizes are character counts of the article source. The signed number is the change from the previous revision.
Recent edit summaries
Detailed summaries recorded by editors. Generic maintenance summaries are omitted here; every recorded revision remains below. Dates describe the edit, not necessarily the event it covers.
- Correction: Corrected the 2-bit OPT-175B perplexity figures, the group-quantization storage overhead, the GPU memory of the RTX 3090/4090, the claim that 3-bit RTN only collapses below 30B, TheBloke's name (Tom Jobbins), the description of the MARLIN paper, the expansion of AQLM, and the source/hardware of the vLLM Marlin throughput benchmark (plus 21 minor fixes)
Version 6 · Nov 4, 2026, 07:19 PM
- Improve internal links (2026-10-13)
Version 5 · Oct 13, 2026, 10:20 AM
- Added 6 contextual internal links
Version 4 · Jul 23, 2026, 03:29 PM