本地大模型部署 A high-throughput and memory-efficient inference and serving engine for LLMs
Python · active 2025-05-04 → 2026-08-10 (UTC)
Activity over time
Daily event counts in the loaded window
Line chart, 464 days from 2025-05-04 to 2026-08-10. Pushes: 7,983 total, peak 51 in a day. Pull requests: 19,769 total, peak 152 in a day. Issues: 8,677 total, peak 67 in a day. Comments: 65,907 total, peak 456 in a day. Stars: 13,377 total, peak 115 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 |
|---|---|---|---|---|
| gemini-code-assist[bot] | 12.9K | 0 | 0 | 6.8K |
| mergify[bot] | 11.9K | 18 | 7K | 4.8K |
| github-actions[bot] | 11.6K | 0 | 605 | 7.7K |
| DarkLight1337 | 8.5K | 883 | 784 | 3.7K |
| mgoin | 2.8K | 385 | 577 | 849 |
| ProExpertProg | 2.5K | 143 | 105 | 1.3K |
| njhill | 2.3K | 169 | 257 | 1.1K |
| vllm-bot | 2.2K | 1.4K | 667 | 0 |
| hmellor | 2.2K | 133 | 231 | 1.1K |
| yewentao256 | 1.7K | 455 | 218 | 432 |
| robertgshaw2-redhat | 1.7K | 524 | 117 | 554 |
| Isotr0py | 1.5K | 179 | 201 | 570 |
| WoosukKwon | 1.5K | 659 | 288 | 285 |
| noooop | 1.3K | 74 | 96 | 786 |
| LucasWilkinson | 1.3K | 197 | 169 | 506 |
| chatgpt-codex-connector[bot] | 1.3K | 0 | 0 | 772 |
| jeejeelee | 1.3K | 125 | 168 | 516 |
| heheda12345 | 1.2K | 67 | 82 | 657 |
| tjtanaa | 1.1K | 160 | 115 | 455 |
| AndreasKaratzas | 1.1K | 60 | 200 | 524 |
Recent activity
Latest issues, pull requests and releases
- Issue comment#51659nebius-llm-robot2026-08-10 11:23Add opt-in persistent torch compile cache
- Pull request#51657hmellor2026-08-10 10:41
- Pull request#51511ciru-ai2026-08-08 18:19
- Pull request#51490mayuriphad2026-08-08 07:40
- Issue comment#47205github-actions[bot]2026-08-08 06:18[Kernel][XPU] Tensor-descriptor operand loads for Triton W8A8 scaled_mm
- Issue comment#51432DarkLight13372026-08-08 04:08[Bugfix][multi_modal] Fix pos_ids being unitialized for minicpmv2.6 in hf runner
- Pull request#51185Harry-Chen2026-08-08 04:07
- Issue comment#51076github-actions[bot]2026-08-08 03:39[Bugfix][Multimodal] Fix PyNvVideoCodec video backend returning NCHW instead of NHWC
- Issue comment#51076Isotr0py2026-08-08 03:38[Bugfix][Multimodal] Fix PyNvVideoCodec video backend returning NCHW instead of NHWC
- Pull request#51474mergify[bot]2026-08-08 03:06
- Issue comment#48003github-actions[bot]2026-08-08 03:06[Bugfix] Fix corrupted output with FlashInfer + decode context parallelism
- Issue comment#50870686f6c612026-08-07 22:05[BugFix][Kernel] Pass kv_cache_dtype as int to avoid stable-ABI host leak
- Issue comment#50870mergify[bot]2026-08-07 21:23[BugFix][Kernel] Pass kv_cache_dtype as int to avoid stable-ABI host leak
- Pull request#51453giuseppegrossi2026-08-07 21:22
- Issue comment#51440yewentao2562026-08-07 21:15[CI Test] Add specific unit test for mrv2 offloading
- Pull request#51410github-actions[bot]2026-08-07 20:37
- Pull request#51440yewentao2562026-08-07 20:03
- Issue comment#49328github-actions[bot]2026-08-07 19:50[KV Offload] Fix failed-load livelock by marking the lookup verdict as a miss
- Pull request#51438mergify[bot]2026-08-07 19:46
- Issue comment#51432github-actions[bot]2026-08-07 19:42[Bugfix][multi_modal] Fix pos_ids being unitialized for minicpmv2.6 in hf runner
- Pull request#51435mergify[bot]2026-08-07 19:35
- Pull request#50484WoosukKwon2026-08-07 19:32
- Pull request#50062TheEpicDolphin2026-08-07 19:16
- Pull request#50930aarushjain292026-08-07 19:07
- Pull request#51433wseaton2026-08-07 18:56
Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 13,377 stars here means stars gained during the window, not the repo's star count.