MSCCL++: A GPU-driven communication stack for scalable AI applications
active 2023-11-22 → 2026-08-13 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 996 days from 2023-11-22 to 2026-08-13. Pushes: 2,358 total, peak 27 in a day. Pull requests: 470 total, peak 8 in a day. Issues: 101 total, peak 8 in a day. Comments: 1,173 total, peak 23 in a day. Stars: 292 total, peak 34 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
| Contributor | Contributions | Pushes | PRs | Comments |
|---|---|---|---|---|
| Binyang2014 | 1.7K | 1K | 121 | 313 |
| chhwang | 1.3K | 595 | 205 | 282 |
| caiomcbr | 455 | 264 | 63 | 55 |
| Copilot | 358 | 26 | 18 | 172 |
| seagater | 292 | 153 | 9 | 67 |
| azure-pipelines[bot] | 128 | 0 | 0 | 128 |
| Empyreus | 80 | 80 | 0 | 0 |
| mahdiehghazim | 76 | 45 | 5 | 18 |
| saeedmaleki | 66 | 44 | 1 | 9 |
| jeseszhang1010 | 42 | 42 | 0 | 0 |
| liangyuRain | 41 | 0 | 2 | 24 |
| roshandathathri | 34 | 11 | 7 | 8 |
| yzygitzh | 32 | 3 | 9 | 8 |
| copilot-pull-request-reviewer[bot] | 26 | 0 | 0 | 4 |
| SreevatsaAnantharamu | 22 | 17 | 2 | 1 |
| FC-Li | 16 | 0 | 0 | 11 |
| cp5555 | 12 | 0 | 0 | 5 |
| microsoft-github-policy-service[bot] | 10 | 0 | 0 | 10 |
| chenhongyu2048 | 9 | 0 | 0 | 5 |
| pash-msft | 7 | 2 | 4 | 0 |
Recent activity
Latest issues, pull requests and releases
- Pull request#864chhwang2026-08-07 15:57
- Issue comment#798azure-pipelines[bot]2026-05-18 20:56Multinode NVL/NVLS supports
- Pull request#805chhwang2026-05-15 22:52
- Pull request#802chhwang2026-05-13 17:11
- Pull request#795Binyang20142026-05-02 03:57
- Issue comment#793azure-pipelines[bot]2026-04-26 07:02GDRCopy status message to string
- Issue comment#792Binyang20142026-04-24 23:04Fix FP8 ROCm build/test issues and dtype naming
- Pull request#792Binyang20142026-04-24 22:17
- Issue#788Binyang20142026-04-22 17:12"RuntimeError: cuMemMap is used in env without NVLS support (mscclpp failure: InvalidUsage)" Error
- Pull request#790Copilot2026-04-22 17:12
- Pull request#790Copilot2026-04-20 00:32
- Pull request#790Copilot2026-04-20 00:32
- Issue comment#787azure-pipelines[bot]2026-04-15 23:56Support python wheel build
- Issue comment#786azure-pipelines[bot]2026-04-15 03:32Fix nccl-test CI building for all GPU architectures
- Pull request#781Binyang20142026-04-14 04:51
- Pull request#783caiomcbr2026-04-11 21:23
- Issue#772SetsuY2026-04-10 19:55[Bug] std::bad_cast on torch integration
- Issue comment#772SetsuY2026-04-10 19:55[Bug] std::bad_cast on torch integration
- Issue comment#781azure-pipelines[bot]2026-04-10 17:42Fix multi-node H100 CI: CUDA compat, deploy improvements
- Pull request#777caiomcbr2026-04-08 23:03
- Pull request#776seagater2026-04-08 21:59
- Issue comment#753chhwang2026-04-08 00:43IB `host-no-atomic`: GDRCopy + mlx5dv Data Direct for memory-consistent low-latency signaling
- Issue comment#774azure-pipelines[bot]2026-04-07 23:28Add ROCm FP8 E4M3B15 support
- Issue comment#774Binyang20142026-04-07 23:27Add ROCm FP8 E4M3B15 support
- Issue comment#774azure-pipelines[bot]2026-04-07 22:46Add ROCm FP8 E4M3B15 support
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 292 stars here means stars gained during the window, not the repo's star count.