mirror of
https://github.com/langgenius/dify-official-plugins.git
synced 2026-07-21 17:45:23 -04:00
OpenAI-API-compatible add qwen3-235b-a22b error #318
Reference in New Issue
Block a user
Delete Branch "%!s()"
Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?
Originally created by @derekmaoo on GitHub (May 27, 2025).
Self Checks
Dify version
1.2.0
Cloud or Self Hosted
Self Hosted (Docker)
Steps to reproduce
1.Add config in model supplier
2.config OpenAI-API-compatible
3.add model
4.config qwen3-235b-a22b
✔️ Expected Behavior
config success
❌ Actual Behavior
Credentials validation failed with status code 400
@dosubot[bot] commented on GitHub (May 27, 2025):
Hi @derekmaoo! I'm Dosu and I’m helping the dify team.
The "Credentials validation failed with status code 400" error when adding the
qwen3-235b-a22bmodel via the OpenAI-API-compatible interface in Dify is most often caused by one of the following issues:Parameter mismatch: For some models, Dify expects the parameter
max_completion_tokensinstead ofmax_tokens. If your configuration usesmax_tokens, try changing it tomax_completion_tokensand see if the error resolves.Reference
Provider selection: Users have reported that using the "OpenAI" provider directly (instead of the "OpenAI API Compatible" provider) resolves this error for custom models.
Reference
Credential configuration: Make sure you have entered a valid API key in the model provider settings. The credential schema for OpenAI-compatible models requires an
openai_api_key(mandatory), and optionallyopenai_organizationandopenai_api_base. Double-check that your API key is correct and does not contain any extra or hidden characters.Reference
Model name formatting: Ensure that your model name does not start with a
/, as this can also trigger the 400 error.Reference
General troubleshooting: If you are configuring the model for the first time, try refreshing the page, re-entering the model and API key, and avoid changing other parameters. Sometimes, input issues (like hidden characters) can cause credential validation to fail.
Reference
Summary of steps to resolve:
max_completion_tokensinstead ofmax_tokensin your config./.If you follow these steps and still encounter issues, please share your exact configuration (with sensitive info redacted) for further troubleshooting.
To reply, just mention my name @dosu.
How did I do? Good | Irrelevant | Incorrect | Verbose | Hallucination | Report 🐛 | Other

@mmobai commented on GitHub (Jun 11, 2025):
same as you ,do you fix it?
@derekmaoo commented on GitHub (Jun 13, 2025):
Main causes of the problem:
There is a rule restriction on the general sense side,enable_thinking must be set to false for non-streaming calls, hope the community can solve it as soon as possible
The Solution:
@dosubot[bot] commented on GitHub (Aug 20, 2025):
Hi, @derekmaoo. I'm Dosu, and I'm helping the dify-official-plugins team manage their backlog and am marking this issue as stale.
Issue Summary:
enable_thinkingto be false for non-streaming calls.max_completion_tokensinstead ofmax_tokens, switching provider to "OpenAI," verifying API key, and avoiding model names starting with "/".Next Steps:
Thanks for your understanding and contribution!