Skip to content

Official implementation of Pref-GRPO: Pairwise Preference Reward-based GRPO for Stable Text-to-Image Reinforcement Learning

active 2025-08-292026-05-09 (UTC)

Partial coverage12,655 / 14,401 hourly files (88%) · 2 absent upstream · 1,740 failed, retryable2024-12-192026-08-11 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
207
Pushes
42
Pull requests
0
Issues
16
Stars
128
Forks
3

Activity over time

Daily event counts in the loaded window

Line chart, 254 days from 2025-08-29 to 2026-05-09. Pushes: 42 total, peak 4 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 16 total, peak 2 in a day. Comments: 18 total, peak 6 in a day. Stars: 128 total, peak 47 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
CodeGoat246042013
Reuben-Sun4002
xxheyu2000
Shelsin2001
yuzhang3006242001
wikeeyang2001
xingzhejun1000
MengHao6661000
Lxiangyue1000
NielsRogge1000

Recent activity

Latest issues, pull requests and releases

  • Issue#4CodeGoat242025-11-29 18:58
    any plan or report for video?
  • Issue comment#4CodeGoat242025-11-29 18:58
    any plan or report for video?
  • Issue#14CodeGoat242025-11-28 19:21
    What does the reward curve during training look like?
  • Issue#12CodeGoat242025-11-28 19:21
    will u support qwen-image ?
  • Issue comment#12CodeGoat242025-10-25 04:55
    will u support qwen-image ?
  • Issue comment#13CodeGoat242025-10-18 14:11
    Pretrained model checkpoints?
  • Issue#12MengHao6662025-10-17 02:14
    will u support qwen-image ?
  • Issue comment#2Shelsin2025-09-18 05:49
    Reward hacking
  • Issue#8CodeGoat242025-09-18 05:48
    About the training time
  • Issue comment#11CodeGoat242025-09-18 04:08
    分数使用问题
  • Issue#10Reuben-Sun2025-09-15 08:58
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue comment#10Reuben-Sun2025-09-14 14:11
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue comment#1CodeGoat242025-09-14 14:06
    Training code questions
  • Issue comment#3CodeGoat242025-09-14 14:03
    Release UniGenBench dataset on Hugging Face
  • Issue#3CodeGoat242025-09-14 14:03
    Release UniGenBench dataset on Hugging Face
  • Issue comment#10CodeGoat242025-09-14 13:59
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue comment#10Reuben-Sun2025-09-14 13:43
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue comment#10CodeGoat242025-09-14 09:51
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue#10Reuben-Sun2025-09-14 07:04
    ModuleNotFoundError: No module named 'fastvideo.distill'
  • Issue#1xxheyu2025-09-10 10:58
    Training code questions
  • Issue comment#9yuzhang3006242025-09-10 06:33
    请问如何判断模型出现奖励骇客
  • Issue#9yuzhang3006242025-09-10 06:33
    请问如何判断模型出现奖励骇客
  • Issue comment#9CodeGoat242025-09-10 02:15
    请问如何判断模型出现奖励骇客
  • Issue#8Lxiangyue2025-09-04 04:53
    About the training time
  • Issue comment#6CodeGoat242025-09-03 09:14
    请教下flow_step和dance_step的区别

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 128 stars here means stars gained during the window, not the repo's star count.