Pulse Contributors Code Frequency Recent Commits

2026-07-24 - 2026-10-24

Period: 3 months

1 day 3 days 1 week 1 month 3 months 6 months 1 year

Overview

2 Active Pull Requests

4 Active Issues

0
Merged Pull Requests 2
Proposed Pull Requests 0
Closed Issues 4
New Issues

Excluding merges, 279 authors have pushed 1164 commits to main and 1729 commits to all branches. On main, 3865 files have changed and there have been 784095 additions and 184592 deletions.

9 Releases published by 1 user

Published v0.31.0rc1 2026-10-05 22:10:24 -04:00

Published v0.30.0 2026-09-25 13:42:53 -04:00

Published v0.30.0rc1 2026-09-22 11:52:01 -04:00

Published v0.29.0rc1 2026-09-10 00:08:28 -04:00

Published v0.28.0 2026-08-31 11:27:13 -04:00

Published v0.28.0rc1 2026-08-27 02:12:55 -04:00

Published v0.27.0rc1 2026-08-11 11:23:46 -04:00

Published v0.26.0 2026-08-03 04:24:50 -04:00

Published v0.26.0rc1 2026-07-27 19:35:35 -04:00

2 Pull requests proposed by 1 user

Proposed #4 probe: Qwen-Image conditioning-KV spill probe on RTX 3090 / RTX 5080 (issue 2) 2026-10-08 00:58:57 -04:00

Proposed #6 Int8 weight-only compressed-tensors DiT artifact + serve-path fixes (issue 5) 2026-10-08 08:34:15 -04:00

4 Issues created by 1 user

Opened #1 Validate Qwen-Image-2.1 serving on Ampere (RTX 3090) and report upstream 2026-10-07 18:16:57 -04:00

Opened #2 Probe whether the Qwen-Image conditioning KV can spill to host DRAM on the RTX 3090 2026-10-07 21:02:38 -04:00

Opened #3 Investigate and fix the TP chroma-speckle corruption upstream (vllm-omni#8135) 2026-10-08 00:08:21 -04:00

Opened #5 Serve a pre-quantized int8 weight-only DiT via compressed-tensors for 3-4 reference capacity on the RTX 3090 2026-10-08 06:34:36 -04:00