Yeuoly comments

Results 139 comments of


                                            Yeuoly

OpenAI-API-compatible model can't support Perplexity api

Pls specify the delimiter for streaming result to "\n" instead of "\n\n" by default while setup credentials, this issue is due to pplx api is not a standard OpenAI compatible...

feat: add support for request timeout settings in the HTTP request node.

I suggest a max timeout to avoid network blocking issues, otherwise it brings a memory leak problem.

feat: add support for request timeout settings in the HTTP request node.

> > I suggest a max timeout to avoid network blocking issues, otherwise it brings a memory leak problem. > > The timeout period can vary depending on different business...

add custom tool with openapi protocol failed

sorry for the bad experience, however, object type is not completely supported currently, we will enhance this part in month, thx for your understanding.

Code execution service is unavailable

Are you using sandbox at high frequency? just try to increase MAX_WORKERS in environment of sandbox service.

Multi-turn conversation that involve images are not supported when using the gpt-4-turbo model.

> I noticed in the source code that when using the `gpt-4-turbo` and `gpt-4-turbo-2024-04-09` models, the user-uploaded images are replaced with the text `[IMAGE]`. But I don't understand why this...

fix: delete tool parameters cache when sync draft workflow for run workflow use new parameter change in draft workflow

Thx!

fix: delete tool parameters cache when sync draft workflow for run workflow use new parameter change in draft workflow

Pls fix the CI error.

The code execution node in the workflow reports an error when parsing the knowledge base output result

Could you pls provide a bit more detailed information? such as the code.

If llm prompt contains:`<script>`, the workflow cannot run

I cannot reproduce this issue, maybe you can share your DSL so that I can locate problems quickly, thx!