dvlab-research/MGM
Official repo for "Mini-Gemini: Mining the Potential of Multi-modality Vision Language Models"
PythonApache-2.0
Issues
- 4
为啥我6张4090 24G,微调Mini-Gemini-8x7B会显存不够QAQ
#75 opened by HongLouyemeng - 4
可以放一下生成generation_pure_text数据的代码吗
#109 opened by pennypengpm - 1
Generation-related Instructions dataset link
#112 opened by berry-ding - 0
关于多机多卡效果不如单机多卡好的问题
#111 opened by DePengW - 0
Loss does not decrease
#110 opened by yfthu - 1
- 10
请问为什么在执行”python -m minigemini.serve.controller --host 0.0.0.0 --port 10000“时会出现 404 Not Found
#76 opened by liuwenxin0410 - 3
- 2
Which deepspeed version is it
#91 opened by Kareneveve - 2
- 0
LLama 70B support
#108 opened by PrateekPal641 - 0
Inference speed
#107 opened by PrateekPal641 - 2
Some weights of the model checkpoint were not used when initializing MGMLlamaForCausalLM
#103 opened by charlesCXK - 1
lora initialisation missing from builder.py
#106 opened by adrielkuek - 1
how to use stage2 ckpt fine-tuning stage3?
#102 opened by linqinguang - 1
- 1
Excessive Length of Responses from Mini Gemini
#97 opened by Dopplenum - 1
Use of ocr in Evaluation
#95 opened by bruceisme - 1
请问为什么在训练llama的脚本中,预训练和微调所使用的conv不一样
#89 opened by shidingz - 1
model asks self questions and answers
#88 opened by Bowei-Li - 1
- 0
计划加入DPO训练来缓解模型幻觉问题吗
#101 opened by jiezhangGt - 3
Some questions about the demo
#85 opened by cyy-1234 - 2
- 2
Take input image as condition.
#100 opened by Adenialzz - 1
stage2 loss is 0
#98 opened by jiezhangGt - 15
- 1
Deployed mini-Gemini in the Windows system and encountered the following error during the ”Launch a Graph web server“ step. seeking help from a skilled user to resolve the issue
#81 opened by Janusmsr - 1
- 2
当我使用推理命令的时候出现网络错误,无法构建推理的接口
#90 opened by HongLouyemeng - 2
how to prompt to get short response
#86 opened by AllenDun - 1
Huggingface inference script
#84 opened by berry-ding - 7
Finetune
#83 opened by ZhangScream - 0
为什么输出结果为nan呢
#87 opened by freja-zy - 4
部署成功试了后,有时会循环输出,还有对中文不是很友好
#65 opened by chenhaoqiang - 2
Deployed mini-Gemini in the Windows system and encountered the following error during the “Launch a Graph web server” step.Seeking help to resolve the issue
#82 opened by Janusmsr - 1
- 3
- 1
Can this model do graph prediction tasks? For example, predict the future trend of personal social graph.
#79 opened by brainplait - 1
ModuleNotFoundError: No module named 'open_clip'
#74 opened by XHB-ZMM - 1
You are using a model of type mini_gemini_mixtral to instantiate a model of type mini_gemini. This is not supported for all configurations of models and can yield errors.
#63 opened by lightingvector - 1
- 1
如何调用api
#67 opened by RoronoaZoroh - 4
- 2
Inquery about simple request
#72 opened by madhatter349 - 0
Inquery about the missing images from ocr_vqa, sam, gpt4v-dataset and ALLaVA-4V
#71 opened by patrick-tssn - 0
button and menu clickdown does not work
#70 opened by jeevikasirwani - 0
add autoscroll
#69 opened by jeevikasirwani - 2
Failed to continous sft for yi-34B with 8x CUDA graphics card! (deepspeed zero3)
#59 opened by xylcbd - 4