Skip to content

3D-R1: Enhancing Reasoning in 3D VLMs for Unified Scene Understanding

active 2025-07-302026-04-20 (UTC)

Partial coverage11,692 / 12,932 hourly files (90%) · 2 absent upstream · 1,236 failed, retryable2025-02-182026-08-10 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
372
Pushes
45
Pull requests
0
Issues
40
Stars
222
Forks
6

Activity over time

Daily event counts in the loaded window

Line chart, 265 days from 2025-07-30 to 2026-04-20. Pushes: 45 total, peak 8 in a day. Pull requests: 0 total, peak 0 in a day. Issues: 40 total, peak 4 in a day. Comments: 59 total, peak 6 in a day. Stars: 222 total, peak 27 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

Recent activity

Latest issues, pull requests and releases

  • Issue#47Believeht0292026-02-18 09:25
    custom data inference
  • Issue#49tengchenzeal2026-02-05 07:53
    关于使用RGB、深度图输入运行3D-R1 checkpoint的问题
  • Issue comment#48tengchenzeal2026-01-30 07:18
    About point cloud encoder
  • Issue#48tengchenzeal2026-01-30 03:46
    About point cloud encoder
  • Issue#47Archer4922025-12-30 06:41
    custom data inference
  • Issue#46lbn020124-droid2025-12-26 11:43
    Data download
  • Issue comment#32Believeht0292025-12-22 05:28
    some worry
  • Issue#44Believeht0292025-12-17 04:17
    Potential Data Leakage: Validation scenes seems included in SFT Cold Start Dataset in `scenecold_dataset.py`
  • Issue#45Believeht0292025-12-17 04:16
    Installation of depthanything
  • Issue comment#25liuQuan982025-12-08 22:37
    Provided Code Not Reproducible — Multiple Training and Evaluation Failures
  • Issue#40Believeht0292025-12-05 04:16
    loss non-finite during SFT training
  • Issue#42Believeht0292025-12-01 02:59
    build the Scene-30k dataset
  • Issue comment#42Tttttt3052025-12-01 02:58
    build the Scene-30k dataset
  • Issue comment#40Believeht0292025-11-24 16:31
    loss non-finite during SFT training
  • Issue comment#25sheyuew2025-11-24 09:12
    Provided Code Not Reproducible — Multiple Training and Evaluation Failures
  • Issue#36Believeht0292025-11-14 14:24
    Acquisition of external multi-view and depth maps
  • Issue comment#35wolf9431344972025-11-12 08:33
    3D-R1 custom data inference demo
  • Issue comment#35Believeht0292025-11-10 08:17
    3D-R1 custom data inference demo
  • Issue comment#36Believeht0292025-11-10 06:37
    Acquisition of external multi-view and depth maps
  • Issue comment#36Believeht0292025-11-10 06:34
    Acquisition of external multi-view and depth maps
  • Issue#1Believeht0292025-11-10 06:14
    Question about visualization
  • Issue comment#25Believeht0292025-11-02 07:16
    Provided Code Not Reproducible — Multiple Training and Evaluation Failures
  • Issue comment#25Believeht0292025-11-02 05:13
    Provided Code Not Reproducible — Multiple Training and Evaluation Failures
  • Issue comment#25Djerry-h2025-11-02 04:23
    Provided Code Not Reproducible — Multiple Training and Evaluation Failures
  • Issue#34wolf9431344972025-10-31 04:35
    More details about environment setup

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 222 stars here means stars gained during the window, not the repo's star count.