fix: correct model type extraction for O1 model handling

[mrT23]

User description

#1294

---

PR Type

bug_fix

---

Description

• Fixed the extraction of the model type to correctly handle Azure O1 models by splitting the model string.
• Adjusted the logic to combine system and user prompts for models with the O1 prefix.
• This change addresses an issue where the system role was not supported by certain Azure models.

---

Changes walkthrough 📝

	Relevant files
Bug fix	litellm_ai_handler.py Fix model type extraction and prompt handling for O1 models pr_agent/algo/ai_handlers/litellm_ai_handler.py Extracted model type from the model string using split('/'). Adjusted condition to check if the model type starts with the O1 model prefix. Combined system and user prompts for O1 models. +2/-1

---

💡 PR-Agent usage: Comment /help "your question" on any pull request to receive relevant information

[codiumai-pr-agent-pro]

PR Reviewer Guide 🔍

(Review updated until commit dcb7b66)

Here are some key observations to aid the review process:

🎫 Ticket compliance analysis ✅ 1294 - Fully compliant Fully compliant requirements: Fix the extraction of the model type to correctly handle Azure O1 models Adjust the logic to combine system and user prompts for models with the O1 prefix Address the issue where the system role was not supported by certain Azure models

⏱️ Estimated effort to review: 2 🔵🔵⚪⚪⚪

🏅 Score: 90

🧪 No relevant tests

🔒 No security concerns identified

🔀 No multiple PR themes

⚡ Recommended focus areas for review Potential Bug The new code assumes that the model string always contains a '/', which might not be the case. Consider adding a check for this scenario.

[mrT23]

Persistent review updated to latest commit dcb7b66

[mrT23]

PR Code Suggestions ✨

Latest suggestions up to dcb7b66

Category	Suggestion	Score
Enhancement	Use a more robust method to extract the model type from the model path Consider using the os.path.split() function from the os.path module to safely extract the model type from the model path. This approach is more robust and works correctly on different operating systems. pr_agent/algo/ai_handlers/litellm_ai_handler.py [191] -model_type = model.split('/')[-1] if '/' in model else model +import os +model_type = os.path.split(model)[1] Apply this suggestion Suggestion importance[1-10]: 8 Why: The suggestion to use os.path.split() is more robust and works correctly across different operating systems, which is a significant improvement over the current method.	8
Possible issue	Add a check for an empty model string before extracting the model type Consider adding a check to ensure that the model string is not empty before attempting to extract the model type. This can prevent potential errors if an empty string is passed as a model. pr_agent/algo/ai_handlers/litellm_ai_handler.py [191-192] -model_type = model.split('/')[-1] if '/' in model else model -if model_type.startswith(O1_MODEL_PREFIX): +if model: + model_type = model.split('/')[-1] if '/' in model else model + if model_type.startswith(O1_MODEL_PREFIX): Apply this suggestion Suggestion importance[1-10]: 7 Why: Adding a check for an empty model string is a good practice to prevent potential errors, enhancing the robustness of the code.	7
Best practice	Use a constant for the separator character to improve code maintainability Consider using a constant for the separator character ('/') to improve code readability and maintainability. This makes it easier to update the separator if needed in the future. pr_agent/algo/ai_handlers/litellm_ai_handler.py [191] -model_type = model.split('/')[-1] if '/' in model else model +SEPARATOR = '/' +model_type = model.split(SEPARATOR)[-1] if SEPARATOR in model else model Apply this suggestion Suggestion importance[1-10]: 6 Why: Using a constant for the separator character improves code readability and maintainability, making it easier to update the separator if needed in the future.	6

• Author self-review: I have reviewed the PR code suggestions, and addressed the relevant ones.

---

Previous suggestions

✅ Suggestions up to commit b743714

Category	Suggestion	Score
Maintainability	Add logging for extracted model type to improve debugging capabilities Consider adding a log message when the model type is extracted to aid in debugging and monitoring. pr_agent/algo/ai_handlers/litellm_ai_handler.py [191-192] model_type = model.split('/')[-1] # 'azure/o1-' or 'o1-' +get_logger().debug(f"Extracted model type: {model_type}") if model_type.startswith(O1_MODEL_PREFIX): Suggestion importance[1-10]: 6 Why: Adding a debug log for the extracted model type can aid in monitoring and debugging, providing visibility into the model type being processed. This is a useful enhancement for maintainability and debugging.	6
Enhancement	✅ Improve the robustness of model type extraction to handle various input formats Consider using a more robust method to extract the model type, such as regex or string manipulation, to handle potential variations in model naming conventions. pr_agent/algo/ai_handlers/litellm_ai_handler.py [191-192] -model_type = model.split('/')[-1] # 'azure/o1-' or 'o1-' +model_type = model.split('/')[-1] if '/' in model else model if model_type.startswith(O1_MODEL_PREFIX): [Suggestion has been applied] Suggestion importance[1-10]: 5 Why: The suggestion improves the robustness of model type extraction by handling cases where the model string may not contain a '/', which could prevent potential errors. However, it does not fully address the suggestion content's mention of using regex or more advanced string manipulation techniques.	5