Slow Embedding Performance with External Model API Calls in Dify - Need for Concurrency Adjustment

[hw-tutu]

Self Checks

• I have searched for existing issues search for existing issues, including closed ones.
• I confirm that I am using English to submit this report (我已阅读并同意 Language Policy).
• [FOR CHINESE USERS] 请务必使用英文提交 Issue，否则会被关闭。谢谢！:）
• Please do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing? Tell me about your story.

I have encountered a significant issue with the embedding process.

Specifically, I am using an external embedding model via API calls to convert text into embeddings. Despite the embedding model having sufficient performance to handle the text conversion tasks, the process is extremely slow within Dify.

It was frustrating because the slow embedding process bottlenecks the entire workflow, making it challenging to scale and meet project deadlines. Given that the external embedding model itself is capable, I suspect that the issue might be related to the concurrency settings or limitations within Dify.

2. Additional context or comments

To address this issue, I am seeking information on whether Dify currently supports any concurrency settings for embedding tasks. If such settings exist, please provide guidance on how to configure them to improve performance.

If Dify does not currently offer concurrency settings for embedding tasks, I would like to request the addition of this feature. Enhancing concurrency could significantly improve the efficiency of the embedding process, thereby benefiting all users who rely on external embedding models.

3. Can you help us with this feature?

• I am interested in contributing to this feature.