Skip to content

视觉推理模型

active 2025-02-152026-07-28 (UTC)

Complete coverage26,455 / 26,455 hourly files (100%) · 2 absent upstream2023-08-152026-08-21 (UTC)
Events
6.8K
Pushes
112
Pull requests
93
Issues
374
Stars
5.2K
Forks
330

Activity over time

Daily event counts in the loaded window

Line chart, 529 days from 2025-02-15 to 2026-07-28. Pushes: 112 total, peak 8 in a day. Pull requests: 93 total, peak 9 in a day. Issues: 374 total, peak 19 in a day. Comments: 625 total, peak 41 in a day. Stars: 5,230 total, peak 921 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
SZhanZ214624699
snakeztc1156144
xrc1066271322
zhangqianqianhzlh173104
weizhenli012160014
dulibubai15009
Shengqi77140011
David-1994071812008
zhaoyuxin212005
zilunzhang101000
StinkyTofu95100010
sleepyshep10008
xddun9004
SabaPivot9036
sms-s9004
yannqi8043
Nobitaxi8006
P3ngLiu8314
penciler-star7005
xuliu-cyber7007

Recent activity

Latest issues, pull requests and releases

  • Issue comment#160Wangningsen2026-01-26 08:00
    run lora train: IndexError: pop from an empty deque
  • Issue comment#160llly12342026-01-26 07:47
    run lora train: IndexError: pop from an empty deque
  • Issue comment#353ZhonghangCai2025-12-22 08:30
    ImportError: cannot import name 'Glm4vForConditionalGeneration' from 'transformers'
  • Issue comment#298liyyyyyyyyyyyyy2025-12-21 10:21
    Question about some details of training
  • Issue comment#298liyyyyyyyyyyyyy2025-12-21 10:21
    Question about some details of training
  • Issue comment#353liyyyyyyyyyyyyy2025-12-19 01:45
    ImportError: cannot import name 'Glm4vForConditionalGeneration' from 'transformers'
  • Issue comment#350sleepyshep2025-10-16 07:55
    请问多久可以支持Qwen3-VL
  • Issue comment#350Henry-Lee-real2025-10-16 07:40
    请问多久可以支持Qwen3-VL
  • Issue comment#342Elijah-Yi2025-10-16 03:22
    GUI task multi-image training
  • Issue comment#350sleepyshep2025-10-15 12:49
    请问多久可以支持Qwen3-VL
  • Issue#350Henry-Lee-real2025-10-15 12:25
    请问多久可以支持Qwen3-VL
  • Pull request#349ZiyiTsang2025-10-15 12:03
  • Issue comment#164xiebinghua2025-10-06 11:30
    Memory usage keeps increasing during training
  • Issue comment#242xiebinghua2025-10-06 03:02
    AutoModelForCausalLM Bug: ValueError: Unrecognized configuration class <class 'transformers.models.qwen2_5_vl.configuration_qwen2_5_vl.Qwen2_5_VLConfig'> for this kind of AutoModel: AutoModelForCausalLM.
  • Issue#348xiebinghua2025-10-05 10:09
    您好,能够提供一下具体的,可直接运行的ovd任务的训练脚本?感谢。
  • Issue comment#140sleepyshep2025-09-28 07:35
    About LoRa Resume Training
  • Issue comment#295pengzhansun2025-09-24 12:08
    Lisa test is invalid.
  • Issue comment#295pengzhansun2025-09-24 12:05
    Lisa test is invalid.
  • Issue comment#220ruolinsss2025-09-19 13:26
    bug when using run_grpo_rec_internVL.sh
  • Issue#346zaiquanyang2025-09-19 09:53
    NaN or Inf found in input tensor when using LLaMA-Factory to train the SFT model for REC task.
  • Issue#345wangsujia2025-09-19 07:50
    are you interested using gspo to train a vlm?
  • Issue comment#342jianmanlincjx2025-09-16 07:39
    GUI task multi-image training
  • Issue#343feivellau2025-09-12 06:29
    Why is the reference model not updated during the entire training cycle?
  • Issue#342songzijian19992025-09-11 07:52
    GUI task multi-image training
  • Issue comment#294Aaashair2025-09-08 14:55
    During a certain step in the training process, an NCCL communication timeout was reported, but there was no error message related to any single GPU.

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 5,230 stars here means stars gained during the window, not the repo's star count.