Skip to content

Scalable toolkit for efficient model reinforcement

active 2025-05-162025-06-12 (UTC)

Partial coverage11,206 / 11,874 hourly files (94%) · 2 absent upstream · 665 failed, retryable2025-04-032026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
1.2K
Pushes
438
Pull requests
74
Issues
25
Stars
204
Forks
29

Activity over time

Daily event counts in the loaded window

Line chart, 28 days from 2025-05-16 to 2025-06-12. Pushes: 438 total, peak 32 in a day. Pull requests: 74 total, peak 14 in a day. Issues: 25 total, peak 3 in a day. Comments: 118 total, peak 18 in a day. Stars: 204 total, peak 69 in a day.

  • Pushes
  • Pull requests
  • Issues
  • Comments
  • Stars

Top contributors

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

ContributorContributionsPushesPRsComments
terrykong214882139
parthchadha93371615
SahilJain31489311214
MahanFathi716920
gshennvm463126
yuki-666442507
chtruong814362356
github-merge-queue[bot]303000
yfw292250
hemildesai211304
ashors1191620
jgerh11008
abukharin-nv11333
jiemingz10801
KiddoZhu10403
HeyyyyyyG8620
joyang-nv8800
yaoyu-335500
ahmadki5401
ertkonuk5500

Recent activity

Latest issues, pull requests and releases

  • Pull request#511SahilJain3142025-06-12 20:20
  • Pull request#489chtruong8142025-06-12 13:09
  • Issue#318terrykong2025-06-12 04:07
    Add logging for histogram (over time) of logprob error
  • Pull request#389terrykong2025-06-12 04:07
  • Pull request#510yashaswikarnati2025-06-12 03:35
  • Pull request#509yfw2025-06-12 01:13
  • Issue#508yuki-6662025-06-12 00:33
    support async non-colocated vllm
  • Pull request#504yfw2025-06-11 23:34
  • Pull request#505abukharin-nv2025-06-11 14:30
  • Pull request#503yfw2025-06-11 09:32
  • Issue#501yuki-6662025-06-11 06:49
    Use native ray collective instead of vllm's in non-colocated
  • Pull request#500SahilJain3142025-06-11 00:59
  • Issue comment#322KiddoZhu2025-06-10 23:54
    feat: code execution & tool use
  • Issue#499KiddoZhu2025-06-10 22:51
    Wrong generation length for HFPolicy with stop strings
  • Pull request#495terrykong2025-06-10 21:05
  • Pull request#482parthchadha2025-06-10 05:38
  • Issue comment#301ahmadki2025-06-09 22:25
    attention mask fixes
  • Issue comment#493okuchaiev2025-06-09 17:04
    CUDA OOM when running deepscaler tutorial on 1 A6000
  • Pull request#492MahanFathi2025-06-09 15:47
  • Issue#468terrykong2025-06-09 02:58
    Python 3.10 compatibility broken: `NotRequired` import incompatible with declared `requires-python = ">=3.10"`
  • Pull request#454terrykong2025-06-09 02:58
  • Pull request#490terrykong2025-06-09 01:02
  • Issue#491suhara2025-06-08 01:08
    Add safe_serialization (safetensors) option to the checkpoint conversion script (`examples/convert_dcp_to_hf.py`)
  • Issue comment#481terrykong2025-06-07 06:47
    feat: add nemotron5 sharding
  • Issue comment#404SumitdevelopAI2025-06-07 06:23
    Create RL playbook as a follow up to DAPT

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 204 stars here means stars gained during the window, not the repo's star count.