Choose Model for Autoprompt feature (RULE_CONFIG_GENERATE_TEMPLATE) #2178

Closed
opened 2026-02-21 17:42:38 -05:00 by yindo · 0 comments
Owner

Originally created by @AndyMik90 on GitHub (Apr 12, 2024).

Originally assigned to: @VincePotato on GitHub.

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).
  • Pleas do not modify this template :) and fill in all the required fields.

1. Is this request related to a challenge you're experiencing?

Since some of the models (especially local models) have a hard time to output correct json format for the autoprompt feature (RULE_CONFIG_GENERATE_TEMPLATE prompt), I suggest being able to set the model for this feature by itself in system settings. As now it uses the system LLM, and for some users, that is not the best for creating prompts in correct format.

2. Describe the feature you'd like to see

Set the model for the prompts in the llm_generate, especially the RULE_CONFIG_GENERATE_TEMPLATE one.

3. How will this feature improve your workflow or experience?

It would allow us not to have 400 errors from some models when they are set as the system-wide LLM, and users would always get orchestrated the prompt when using the automatic feature.

4. Additional context or comments

No response

5. Can you help us with this feature?

  • I am interested in contributing to this feature.
Originally created by @AndyMik90 on GitHub (Apr 12, 2024). Originally assigned to: @VincePotato on GitHub. ### Self Checks - [X] I have searched for existing issues [search for existing issues](https://github.com/langgenius/dify/issues), including closed ones. - [X] I confirm that I am using English to submit this report (我已阅读并同意 [Language Policy](https://github.com/langgenius/dify/issues/1542)). - [X] Pleas do not modify this template :) and fill in all the required fields. ### 1. Is this request related to a challenge you're experiencing? Since some of the models (especially local models) have a hard time to output correct json format for the autoprompt feature (RULE_CONFIG_GENERATE_TEMPLATE prompt), I suggest being able to set the model for this feature by itself in system settings. As now it uses the system LLM, and for some users, that is not the best for creating prompts in correct format. ### 2. Describe the feature you'd like to see Set the model for the prompts in the llm_generate, especially the RULE_CONFIG_GENERATE_TEMPLATE one. ### 3. How will this feature improve your workflow or experience? It would allow us not to have 400 errors from some models when they are set as the system-wide LLM, and users would always get orchestrated the prompt when using the automatic feature. ### 4. Additional context or comments _No response_ ### 5. Can you help us with this feature? - [ ] I am interested in contributing to this feature.
yindo added the 💪 enhancementstale labels 2026-02-21 17:42:38 -05:00
yindo closed this issue 2026-02-21 17:42:38 -05:00
Sign in to join this conversation.
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: langgenius/dify#2178