zhanaali
zhanaali
I looked at the code, and it seems to have stopped at this line: "response = dashscope.audio.tts.SpeechSynthesizer.call(model=voice, sample_rate=48000, text=sentence.strip(), format=audio_type)".
Copying the code and running it independently works fine, but I can't get a response inside dify. It's very strange.
There should be no issues with the SDK and the network
[Badcase]: Model inference Qwen2.5-32B-Instruct-GPTQ-Int4 appears as garbled text !!!!!!!!!!!!!!!!!!
Inference file [openai_api_32b.txt](https://github.com/user-attachments/files/17093924/openai_api_32b.txt)
[Badcase]: Model inference Qwen2.5-32B-Instruct-GPTQ-Int4 appears as garbled text !!!!!!!!!!!!!!!!!!
Can the same script reasoning Qwen2.5-32B-Instruct-GPTQ-Int8 model normally output。 Is it a problem with the reasoning parameters?
[Badcase]: Model inference Qwen2.5-32B-Instruct-GPTQ-Int4 appears as garbled text !!!!!!!!!!!!!!!!!!
It's still the same after the upgrade  @hzhwcmhf
@tc-mb 使用的是minicpm-v:2.6的gguf https://ollama.com/library/minicpm-v:8b-2.6-q4_K_M
@tc-mb 你好，最近什么时候会更新