Skip to content

An algorithm for weight-activation quantization (W4A4, W4A8) of LLMs, supporting both static and dynamic quantization

active 2024-10-082025-12-08 (UTC)

Complete coverage26,476 / 26,476 hourly files (100%) · 2 absent upstream2023-08-152026-08-22 (UTC)
Events
255
Pushes
7
Pull requests
0
Issues
39
Stars
144
Forks
8

Activity over time

Daily event counts in the loaded window

Line chart, 427 days from 2024-10-08 to 2025-12-08. Pushes: 7 total, peak 2 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 39 total, peak 5 in a day. Comments: 56 total, peak 5 in a day. Stars: 144 total, peak 11 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Stars, PRs, issues and forks are under-captured in the later part of this window. GH Archive progressively stopped capturing non-push events during 2026 — −95% or worse by the end of the window. Every series here except Pushes fades for that reason, so a decline above reflects the archive, not this repository. Pushes stay reliable throughout, so read them, and the contributor counts derived from them, as the real signal. Data health has the measurements.

Top contributors

Pushes, PRs, issues, reviews and comments — stars and forks excluded, so this is contribution rather than popularity

ContributorContributionsPushesPRsComments
ChenMnZ437028
brisker6005
lonleyodd4002
jameshensman3001
Dianaia3001
yoghur3001
sasha-hailo3002
songh113001
fpcsong3001
L1aoXingyu3002
laomao02001
geqian-91922001
github-yizhang2000
ponytaill2001
yyfcc172001
ostix3602001
z182561992752001
JustVelkhana1000
01000-you1001
lsjlsj58461000

Recent activity

Latest issues, pull requests and releases

  • Issue#31MAI7162025-11-26 06:23
    Bug in get_wikitext2 validation loader in utils/data_utils.py
  • Issue comment#23sasha-hailo2025-10-06 20:24
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue comment#23a192842025-10-06 13:04
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue comment#2454limiao2025-07-18 03:09
    qwen2.5支持吗
  • Issue#27JustVelkhana2025-06-21 05:40
    Can't run evaluate for piqa datasets? May it because there's no lm_eval?
  • Issue comment#25ChenMnZ2025-02-07 12:32
    Question about speedup
  • Issue#25SemyonBevzuk2025-02-07 10:36
    Question about speedup
  • Issue comment#11QingshuiL2025-01-24 02:34
    AttributeError: 'WrappedPrefixCausalLM' object has no attribute 'generate'
  • Issue#24aqe6702025-01-23 11:26
    qwen2.5支持吗
  • Issue comment#23ChenMnZ2025-01-16 02:46
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue comment#19ChenMnZ2025-01-15 02:37
    Not quant Q layer?
  • Issue comment#23ChenMnZ2025-01-15 02:35
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue comment#23sasha-hailo2025-01-14 11:15
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue#23sasha-hailo2025-01-13 11:58
    Possible PrefixQuant issues on Smaller LLMs?
  • Issue comment#22fpcsong2024-12-26 12:22
    Question about rotated model
  • Issue#22fpcsong2024-12-26 12:22
    Question about rotated model
  • Issue#22fpcsong2024-12-26 11:49
    Question about rotated model
  • Issue comment#21laomao02024-12-20 07:07
    Question about Preventing Outlier Tokens during Inference
  • Issue comment#20z182561992752024-12-16 03:08
    How to test chat model
  • Issue comment#20yoghur2024-12-16 02:58
    How to test chat model
  • Issue#20yoghur2024-12-16 02:57
    How to test chat model
  • Issue#20yoghur2024-12-13 10:58
    How to test chat model
  • Issue#19laomao02024-12-13 02:42
    Not quant Q layer?
  • Issue#18z182561992752024-12-12 08:52
    evaluate error
  • Issue#17lsjlsj58462024-12-12 02:29
    Why does finer granularity in quantization result in more overhead?

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 144 stars here means stars gained during the window, not the repo's star count.