Skip to content

PyTorch library for cost-effective, fast and easy serving of MoE models.

active 2024-01-222025-03-03 (UTC)

Complete coverage26,531 / 26,531 hourly files (100%) · 2 absent upstream2023-08-152026-08-24 (UTC)
Events
424
Pushes
97
Pull requests
37
Issues
32
Stars
129
Forks
14

Activity over time

Daily event counts in the loaded window

Line chart, 407 days from 2024-01-22 to 2025-03-03. Pushes: 97 total, peak 7 in a day. Pull requests: 37 total, peak 4 in a day. Issues: 32 total, peak 2 in a day. Comments: 49 total, peak 6 in a day. Stars: 129 total, peak 7 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
drunkcoding142672521
lausannel6926915
future-xy7222
luomai6212
MSGitt3001
JustQJ3001
akhauriyash3001
Msiavashi3002
washing11272001
Tingberer2001
Lincest1001
YLSnowy1000
Hudayday1000
kungfup1000
pegga12251000
nicolay-i1000
alphabewitch1000
taehyunzzz1000
CelestialSteed1001

Recent activity

Latest issues, pull requests and releases

  • Issue#42kungfup2025-03-02 16:15
    FATAL ]ExpertDispatcher::GPUExecFunc DenseActDense tensor_ids: [4 5 6 3 9 10 11 8 14 15 16 13 ] expert_type 4 mat1 and mat2 shapes cannot be multiplied (3x4096 and 14336x1)
  • Pull request#41drunkcoding2025-02-27 02:05
  • Pull request#40drunkcoding2025-02-26 19:39
  • Pull request#40lausannel2025-02-26 18:47
  • Issue comment#39drunkcoding2025-02-23 23:39
    Example of run in google colab
  • Issue comment#33drunkcoding2025-02-23 23:37
    Fail to run readme_example.py
  • Pull request#38drunkcoding2025-02-23 18:14
  • Issue#39nicolay-i2025-02-23 03:24
    Example of run in google colab
  • Issue comment#38drunkcoding2025-02-22 19:08
    Refactor code fr better performance
  • Issue comment#38lausannel2025-02-22 19:00
    Refactor code fr better performance
  • Pull request#38drunkcoding2025-02-22 18:31
  • Issue#23drunkcoding2025-02-19 00:37
    Can the MoE-Infinity framework be used in conjunction with the vLLM framework?
  • Issue#31drunkcoding2025-02-19 00:36
    Question about prefetch implementation.
  • Issue comment#32drunkcoding2025-02-19 00:35
    [Request] Bigger picture of the system
  • Issue#30drunkcoding2025-02-15 15:36
    FATAL Failed to read file: , 314, , errno: , 2, , msg: , No such file or directory
  • Issue#36drunkcoding2025-02-15 15:36
    An error with the example code from README when load model -> ImportError: undefined symbol: _ZN4base6LoggerC1ENS0_10SourceFileEiNS0_8LogLevelEPKc
  • Pull request#37drunkcoding2025-02-15 15:36
  • Pull request#37drunkcoding2025-02-15 15:07
  • Issue comment#36Lincest2025-02-14 08:24
    An error with the example code from README when load model -> ImportError: undefined symbol: _ZN4base6LoggerC1ENS0_10SourceFileEiNS0_8LogLevelEPKc
  • Issue comment#36washing11272025-02-14 03:49
    An error with the example code from README when load model -> ImportError: undefined symbol: _ZN4base6LoggerC1ENS0_10SourceFileEiNS0_8LogLevelEPKc
  • Issue#36washing11272025-02-14 03:41
    An error with the example code from README when load model -> ImportError: undefined symbol: _ZN4base6LoggerC1ENS0_10SourceFileEiNS0_8LogLevelEPKc
  • Pull request#35drunkcoding2025-02-12 02:26
  • Pull request#35lausannel2025-02-11 18:16
  • Pull request#22drunkcoding2025-02-11 16:30
  • Issue#1drunkcoding2025-02-11 16:29
    TODO for first release

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 129 stars here means stars gained during the window, not the repo's star count.