issues
search
OpenBMB
/
MiniCPM-V
MiniCPM-Llama3-V 2.5: A GPT-4V Level Multimodal LLM on Your Phone
Apache License 2.0
7.86k
stars
547
forks
source link
issues
Newest
Newest
Most commented
Recently updated
Oldest
Least commented
Least recently updated
请问lora微调需要多少显存?效果如何?
#327
orderer0001
opened
8 hours ago
0
lora训练请教报错:ImportError: /root/.cache/torch_extensions/py310_cu121/fused_adam/fused_adam.so: cannot open shared object file: No such file or directory 是否环境版本问题
#326
orderer0001
closed
8 hours ago
0
[BUG] <title>3*4090全量微调报错,请教是否参数问题?3*4090 full-scale fine-tuning reports an error. Is there a problem with the parameters?
#325
orderer0001
closed
9 hours ago
0
高精度及多图问题
#324
dingtine
opened
2 days ago
0
💡 [REQUEST] - minicpm是否可以用来做目标检测任务呢?
#323
xyw1
opened
2 days ago
1
finetune代码中patch_attn_mask是不是写错了,运行后全为True了
#322
shidingz
opened
3 days ago
0
不同语言微调的问题💡 [REQUEST] - <title>
#321
y-x-x123
opened
3 days ago
1
finetune on NPUs
#320
EasonXiao-888
opened
3 days ago
0
Where can i find the exact architecture used to train MiniCPM-Llama3-V 2.5 and implementation details
#319
Abhinay2323
opened
4 days ago
2
finetune lora的时候遇到这个问题Input type (torch.cuda.HalfTensor) and weight type (torch.HalfTensor) should be the same
#318
XuanRen4470
opened
5 days ago
0
对于视频数据,除了抽帧再单独处理每一张之外,是否有更好的方法呢?
#317
xyw1
opened
5 days ago
1
[BUG] <为什么在微调代码中的dataset.py 中读取conversation时要remove bos>
#316
Single430
opened
5 days ago
1
vllm推理无法安装环境
#315
iiinsight
closed
5 days ago
1
[BUG] <OCR精度问题>
#314
weiminw
closed
2 days ago
2
what is lr and batch_size in pretrain stage
#313
aoji0606
opened
6 days ago
0
checkpoint shards not loading. Process always gets send to SIGTERM
#312
xsMarc
opened
1 week ago
1
请问,商用的请求已经过去了 7 个工作日,还是没有收到回复的消息
#311
zhjunqin
closed
5 days ago
3
[Issue] AttributeError: 'LoraModel' object has no attribute '_enable_peft_forward_hooks'
#310
ssuncheol
opened
1 week ago
3
[BUG] Pre-training unstable
#309
orrzohar
opened
1 week ago
0
finetuning not working
#308
xsMarc
closed
1 week ago
0
对齐和SFT所用的预训练数据集是什么
#307
AshOneN
opened
1 week ago
0
Whether the model supports batch inference?and whether the relevant example code can be provided?
#306
MrHuangAm
opened
1 week ago
2
[BUG] 你好,我想问下训练数据是图片的话,坐标里的xy对应的是图片的宽高吗
#305
leohl1997
opened
1 week ago
1
Update requirements.txt for finetuning requirements
#304
HelloWorldLTY
opened
1 week ago
0
如何使得MiniCPM-Llama3-V-2.5识别图片中是否存在某个图案
#303
mobguang
opened
1 week ago
7
OpenBMB
#302
kronic9119
opened
1 week ago
0
Clear the torch cuda cache after response
#301
RandomGitUser321
opened
1 week ago
0
[Data info] MiniCPM-llama3-V 2.5
#300
emanuelevivoli
opened
1 week ago
0
[BUG] <title>out of memory
#299
limllzu
closed
1 week ago
0
Lora微调后要怎么加载测试呢?
#298
qazqaz12378
opened
1 week ago
18
微信二维码过期,帮忙更新,谢谢![BUG] <title>
#297
baleli668
closed
1 week ago
1
Special tokens have been added in the vocabulary, make sure the associated word embeddings are fine-tuned or trained. 怎么解决
#296
Zhaojjjjjj
closed
1 week ago
0
[BUG] <title>Can't the chat method pass in the max_tokens or max_new_tokens parameters? What if the length is managed?
#295
orderer0001
opened
1 week ago
1
[BUG] Mac MiniCPM-Llama3-V-2_5 推理出错
#294
mobguang
closed
1 week ago
1
Update inference_on_multiple_gpus.md
#293
glenn124f
opened
2 weeks ago
0
[BUG] 双卡运行device='auto'后NotImplementedError: Cannot copy out of meta tensor; no data!”
#292
bailove
closed
2 weeks ago
2
<|reserved_special_token|>的作用?
#291
ZHANGTIANYAO1
opened
2 weeks ago
0
minicpm-llama3-v2_5模型微调 model_max_length参数
#290
todaydeath
opened
2 weeks ago
1
[BUG] minicpm-llama3-v-2_5(int4)只能使用PIL打开图像吗
#289
morestart
opened
2 weeks ago
0
[BUG] <MiniCPM-Llama3-V-2.5无法在24G 4090上启动>
#288
weiminw
closed
2 weeks ago
13
请问minicpm-llama3-v-2_5可以支持哪些量化措施,支持awq量化吗
#287
1SingleFeng
closed
2 weeks ago
1
MiniCPM-Llama3-V 2.5 全参数微调最小的显存需求
#286
Millie-Xu
opened
2 weeks ago
5
请问跑minicpm-llama3-v-2_5(int4)支持并发调用接口么?2个及以上并发调用就报错了
#285
geminizyz
opened
2 weeks ago
5
有没有大佬能不能发一下跑成功的requirements.txt文件
#284
Fg-droid-1014
closed
2 weeks ago
4
while trying to fine tune model on kaggle this error appear :ValueError: Type fp16 is not supported.
#283
mohamed-em2m
opened
2 weeks ago
1
Question about full-parameter finetuning
#282
dydxdt
opened
2 weeks ago
4
feat: Added judgment logic to support training with plain text data.
#281
hill2hill
opened
2 weeks ago
5
💡 [REQUEST] - <是否有说明可以将该模型发布成openai 适配的api>
#280
weiminw
closed
2 weeks ago
1
Not attention mask is input for training
#279
double-fire-0
closed
2 weeks ago
0
fix a bug with web_demo_streamlit_2.5.py at text mode
#278
RainYooSuki
opened
2 weeks ago
2
Next