-
According https://huggingface.co/spaces/lmsys/chatbot-arena-leaderboard Qwen1.5 model is one of the best OpenSource (Free) models with large context and Russian language support. It would be nice to …
-
### What model would you like?
_No response_
-
### OS
Windows
### GPU Library
CUDA 12.x
### Python version
3.12
### Pytorch version
2.4.0+cu121
### Model
Qwen/Qwen2.5-32B-Instruct
### Describe the bug
Whatever quan…
-
环境:
1.CPU或GPU环境都尝试过了
2.在魔搭社区下载了2-3个LLM模型到本地
场景一:
容器启动CPU镜像:
尝试执行的命令是:docker run --rm --name server --shm-size=50gb -e MODELSCOPE_CACHE=/modelscope_cache -v /root/.cache/modelscope:/mo…
-
我详细的阅读了qwen audio 2的源代码,并对模型的架构进行了进一步的探索。
作者之前声明qwen aduio 2使用的是qwen-1作为llm,但是却在config中出现了qwen2作为text_config,这是令人困惑的。
llm的layer num是32,这与qwen-7b保持一致,但是attention却使用qwen2的attention,让我产生了很大的困惑?
-
**例行检查**
[//]: # (方框内删除已有的空格,填 x 号)
+ [x] 我已确认目前没有类似 issue
+ [x] 我已确认我已升级到最新版本
+ [x] 我已完整查看过项目 README,尤其是常见问题部分
+ [x] 我理解并愿意跟进此 issue,协助测试和提供反馈
+ [x] 我理解并认可上述内容,并理解项目维护者精力有限,**不遵循规则的 issue 可能…
-
Hello,
I am receiving this error:
**_### "An error occurred: The checkpoint you are trying to load has model type `qwen2_vl` but Transformers does not recognize this architecture. This could be be…
-
PC-Agent可以使用国产模型吗?如果可以,代码在哪里改?
-
### Model description
[Qwen/Qwen2-VL-7B-Instruct](https://huggingface.co/Qwen/Qwen2-VL-7B-Instruct)
### Open source status
- [X] The model implementation is available
- [X] The model weights are …
-
It seems like qwen_vl_utils will occupy excessive memory when pre-process, until it get killed by system.
Traceback before it get killed:
```
File "/mnt/workspace/lmms-eval-main/lmms_eval/model…