maxin9966

OpenJTalk undefined symbol: _ZSt28__throw_bad_array_new_lengthv

3

第一次接触这个项目，linux下按照教程部署成功，然后选了一个蔡徐坤的ckpt和pth模型之后，选择运行推理进程，然后就报错了。 "/home/ma/ENTER/envs/tts1/bin/python" GPT_SoVITS/inference_webui.py File "/home/ma/work/GPT-SoVITS/GPT_SoVITS/inference_webui.py", line 64, in from text.cleaner import clean_text File "/home/ma/work/GPT-SoVITS/GPT_SoVITS/text/cleaner.py", line 1, in from text import chinese, japanese, cleaned_text_to_sequence, symbols, english File "/home/ma/work/GPT-SoVITS/GPT_SoVITS/text/japanese.py", line 5, in...

How to automatically generate subtitles?

1

Only can use speech recognition after generation?

Unable to run qwen successfully

15

**env:** 2080Ti * 2 cuda_12.3.r12.3/compiler.33567101_0 python3.9 pip install "sglang[all]" **error:** new fill batch. #seq: 1. #cached_token: 0. #new_token: 8. #remaining_req: 0. #running_req: 0. tree_cache_hit_rate: 0.00%. python: /project/lib/Analysis/Allocation.cpp:40: std::pair mlir::triton::getCvtOrder(mlir::Attribute, mlir::Attribute):...

I intend to use this tool for formatted document layout, as long as it can distinguish between headings of different sizes. Is there a way to achieve this?

[Feature Request]: Why is the SDXL running so slow?

1

### Is there an existing issue for this? - [x] I have searched the existing issues and checked the recent builds/commits ### What would your feature do ? I have...

[BUG] Quantitative model Yi-1.5-9b-16K does not produce text output.

1

**Describe the bug** Quantized the model yi-1.5-9b-16K, quantization was successful, everything runs normally, but no text output is produced. **Hardware details** 4070ti super **Software version** vllm 0.4.2 auto-gptq lastest cuda...

[Feature]: How to run the int4 quantized version of the gemma2-27b model

5

### 🚀 The feature, motivation and pitch How to run the int4 quantized version of the gemma2-27b model ### Alternatives _No response_ ### Additional context _No response_

Calling the API /sdapi/v1/txt2img for text-to-image

9

The API call is /sdapi/v1/txt2img, with the following parameters being passed. The returned image seems to have issues, appearing notably coarser compared to images generated through the web interface. For...

视频理解是否可以使用4bit模型吗？

1

请问视频理解可以使用4bit量化的模型吗？我看文档例子使用的是全量模型，我尝试使用4bit模型提示显存不够，请问一般情况下需要多少显存？ ps: 测试理解单张图像，使用4bit模型是没问题的。 ### 视频理解的代码： from lmdeploy import pipeline, GenerationConfig from transformers.dynamic_module_utils import get_class_from_dynamic_module from lmdeploy import TurbomindEngineConfig, pipeline from lmdeploy.vl import load_image HF_MODEL = 'internlm/internlm-xcomposer2d5-7b' load_video = get_class_from_dynamic_module('ixc_utils.load_video', HF_MODEL)...

[Bug] glm4量化之后，吐字异常重复的问题

6

### Checklist - [x] 1. I have searched related issues but cannot get the expected help. - [x] 2. The bug has not been fixed in the latest version. -...