基于CUDA11.8的 glm4voice 服务(vllm版) 部署参考

[baiyin]

cuda 11.8 版本上搭建 glm4voice，踩了些坑，供参考

https://github.com/baiyin/baiyin.github.io/blob/main/_posts/2024-12-05-deploy-glm4voice-vllm-server-for-cuda118.md

[sixsixcoder]

感谢您的贡献，有问题随时提issues

[wang-TJ-20]

cuda 11.8 版本上搭建 glm4voice，踩了些坑，供参考

https://github.com/baiyin/baiyin.github.io/blob/main/_posts/2024-12-05-deploy-glm4voice-vllm-server-for-cuda118.md

@baiyin ,hi,有试过对cosyvoice2.0进行vllm 加速吗，看到cosyvocie2.0的llm模块更换为了qwen0.5

[changqingla]

请问实测vllm版部署会加快推理速度吗