Skip to content

轻量化大模型(支持图像理解)MiniCPM-Llama3-V 2.5: A GPT-4V Level Multimodal LLM on Your Phone

active 2024-04-112026-08-07 (UTC)

Partial coverage20,517 / 26,273 hourly files (78%) · 2 absent upstream · 5,753 failed, retryable2023-08-152026-08-13 (UTC)— sampled evenly across the window, so rankings and trends hold; absolute counts scale up.
Events
10.9K
Pushes
198
Pull requests
70
Issues
654
Stars
8.1K
Forks
614

Activity over time

Daily event counts in the loaded window

Line chart, 849 days from 2024-04-11 to 2026-08-07. Pushes: 198 total, peak 15 in a day. Pull requests: 70 total, peak 8 in a day. Issues: 654 total, peak 43 in a day. Comments: 1,175 total, peak 35 in a day. Stars: 8,148 total, peak 398 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
qyc-98180171371
Cuiunbo16114349
tc-mb153100108
LDLINGLINGLING147202105
iceflame8912222566
yiranyyu1008833
YuzaChongyi79131434
1SingleFeng250023
todaydeath140013
daihuidai130011
HwwwwwwwH120111
jctime11901
univa-JASON10006
XHB-ZMM9007
orderer00019001
zhaoyang06189008
Stardust-y8007
winsonxxxx8006
ZMXJJ8107
mobguang8006

Recent activity

Latest issues, pull requests and releases

  • Issue comment#1102tc-mb2026-05-18 12:19
    [BUG] 无法加载openbmb/MiniCPM-V-4.6:KeyError: 'minicpmv4_6'
  • Pull request#1098chmod777john2026-05-17 15:47
  • Pull request#1098chmod777john2026-05-15 11:37
  • Issue#1051NiuKequan2026-01-26 20:20
    [BUG] <如何修改 vLLM 离线推理代码以支持minicpm-v-4_5视频推理,有没有示例代码呀>
  • Issue#1050YuXiangLin12342026-01-18 08:31
    [BUG] <title> Minicpm-o-2.6 Audio Tends to Produce Gibberish during Speech-to-Speech Conversations in Audio-Roleplay Mode
  • Issue comment#1049YuzaChongyi2026-01-14 10:44
    [BUG] <title>
  • Issue#10495830543502026-01-14 10:24
    [BUG] <title>
  • Issue comment#1044scut-ken52026-01-09 03:10
    MiniCPM-o 2.6是否能够使用GRPO进行音视频训练?
  • Issue comment#1044tc-mb2026-01-08 08:20
    MiniCPM-o 2.6是否能够使用GRPO进行音视频训练?
  • Issue comment#1044tc-mb2026-01-06 12:02
    MiniCPM-o 2.6是否能够使用GRPO进行音视频训练?
  • Issue comment#1044scut-ken52026-01-06 07:41
    MiniCPM-o 2.6是否能够使用GRPO进行音视频训练?
  • Issue comment#1048tc-mb2026-01-03 08:58
    [BUG] <title>openai.BadRequestError: Error code: 400 - {'error': {'message': 'The decoder prompt (length 2136) is longer than the maximum model length of 2048. }}
  • Issue#1047bihujrj2025-12-19 11:34
    💡 [REQUEST] - <title>只有一个8B版本?
  • Issue comment#1045qyc-982025-12-14 06:35
    evaluation_strategy was renamed to eval_strategy
  • Issue#1045patelddevan2025-12-13 22:17
    evaluation_strategy was renamed to eval_strategy
  • Issue#1043daifuliu2025-12-09 01:46
    [ollama] - OCR not working
  • Issue#1043daifuliu2025-12-09 01:46
    [ollama] - OCR not working
  • Issue#1042rajat-ventura2025-12-06 09:10
    💡 [REQUEST] - Is it possible to train the model with a new language
  • Issue comment#1041tc-mb2025-11-24 03:00
    Getting the time code of a video [bug ?]
  • Issue comment#1040Laxmikant-ArnishAi2025-11-18 09:37
    "Fine-tuning MiniCPM-V-2.6 with QLoRA on a single 12GB GPU fails with TypeError: device() received an invalid combination of arguments - got (NoneType)
  • Issue#1039tc-mb2025-11-14 11:37
    Grounding支持多个目标吗?
  • Issue comment#660tc-mb2025-11-14 11:35
    [BUG] 为什么如何调整参数,显存占用都是接近80G?
  • Issue#660tc-mb2025-11-14 11:35
    [BUG] 为什么如何调整参数,显存占用都是接近80G?
  • Issue#927tc-mb2025-11-14 11:34
    [BUG] <title>flash_attn package
  • Issue#888tc-mb2025-11-14 11:33
    How to use multi modal data with "vllm serve"?

Totals cover only the window loaded into ClickHouse and count events, not GitHub's lifetime totals — 8,148 stars here means stars gained during the window, not the repo's star count.