Skip to content

vllm docker for vLLM on Intel Gaudi

active 2025-06-252026-08-10 (UTC)

Partial coverage11,234 / 11,963 hourly files (94%) · 2 absent upstream · 726 failed, retryable2025-03-302026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
5.5K
Pushes
1.4K
Pull requests
762
Issues
28
Stars
12
Forks
40

Activity over time

Daily event counts in the loaded window

Line chart, 412 days from 2025-06-25 to 2026-08-10. Pushes: 1,358 total, peak 28 in a day. Pull requests: 762 total, peak 31 in a day. Issues: 28 total, peak 4 in a day. Comments: 1,906 total, peak 37 in a day. Stars: 12 total, peak 1 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
github-actions[bot]8011350652
Copilot684276379
xuechendi611176164157
adobrzyn56525762143
PatrykWo2451276333
michalkuligowski244831347
kzawora-intel2331334831
iboiko-habana1971003426
afierka-intel127601919
kamil-kaczor117741221
mgawarkiewicz-intel895845
ksmusz61241312
wpyszka573121
mswiniarsk53181012
yangulei5311620
mhelf-intel494257
skavulya4701119
wuxun-zhang440526
jbyczkow3911241
xinyu-intel3801611

Recent activity

Latest issues, pull requests and releases

  • Issue comment#1566github-actions[bot]2026-06-26 12:21
    Detach shared MoE gate when experts is the MoERunner (vLLM #41184)
  • Pull request#1537yeonsily2026-06-09 22:55
  • Pull request#1536pawel-olejniczak2026-06-09 14:11
  • Pull request#1482rsmyrek2026-05-27 13:52
  • Pull request#1498mkrze2026-05-26 12:19
  • Issue comment#1466github-actions[bot]2026-05-26 09:18
    Fix HPU prompt_token_ids device placement for penalty sampling
  • Issue comment#1494github-actions[bot]2026-05-26 04:19
    Remove transformers instalation from vllm-gaudi
  • Issue comment#1462github-actions[bot]2026-05-26 00:52
    Port of: fix: hybrid model warmup block_size mismatch (Qwen3.5-35B-A3B) #1434
  • Pull request#1491adobrzyn2026-05-25 08:29
  • Pull request#1487osavchenkox2026-05-24 08:53
  • Issue comment#1471github-actions[bot]2026-05-21 10:09
    Add pre-merge-approval for execute_pre_merge
  • Issue comment#1445github-actions[bot]2026-05-20 07:46
    Removal of ray and redundant transformers packages from gaudi requirements
  • Issue comment#1450github-actions[bot]2026-05-15 22:23
    Add Qwen3NextForCausalLM to mamba_like_arch
  • Issue comment#1425yangulei2026-05-14 03:17
    [DOC] Fix torchaudio version
  • Issue comment#1441github-actions[bot]2026-05-13 11:11
    fix: bypass _forward_impl for dp_size==1 to fix DeepSeek R1 FP8 crash
  • Issue comment#1441github-actions[bot]2026-05-13 03:16
    fix: bypass _forward_impl for dp_size==1 to fix DeepSeek R1 FP8 crash
  • Issue comment#1444github-actions[bot]2026-05-13 01:44
    [MiniMax-M2] Remove reduce_results kwarg from FusedMoE init
  • Pull request#1435kamil-kaczor2026-05-12 16:00
  • Issue comment#1347yangulei2026-05-12 00:13
    Supports 256k model length with TP=1 on Gaudi2 for Qwen3-30B-A3B-Thinking-2507
  • Pull request#1433ksmusz2026-05-11 16:01
  • Pull request#1122yangulei2026-05-11 11:57
  • Pull request#1421pawel-olejniczak2026-05-07 07:58
  • Pull request#1409iboiko-habana2026-05-04 09:27
  • Issue comment#1408github-actions[bot]2026-05-02 22:35
    Fix IndexError during decode warmup when seq_length equals max_model_len
  • Issue comment#1402github-actions[bot]2026-04-29 21:58
    To enable defrag if contig_pa is enabled

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 12 stars here means stars gained during the window, not the repo's star count.