Support reasoning effort via configuration

[KennyDizi]

PR Type

Enhancement

---

Description

• Added support for reasoning_effort configuration in AI models.

• Introduced ReasoningEffort enum for reasoning effort levels.

• Updated chat_completion to include reasoning_effort parameter.

• Extended configuration file to allow setting reasoning_effort.

---

Changes walkthrough 📝

	Relevant files
Enhancement	__init__.py Added support for reasoning effort models.                              pr_agent/algo/init.py Defined SUPPORT_REASONING_EFFORT_MODELS list for specific models. +5/-0      litellm_ai_handler.py Enhanced AI handler to support reasoning effort.                  pr_agent/algo/ai_handlers/litellm_ai_handler.py Imported ReasoningEffort and SUPPORT_REASONING_EFFORT_MODELS. Added support_reasoning_models attribute. Updated chat_completion to handle reasoning_effort parameter. +12/-2    utils.py Added ReasoningEffort enum for reasoning levels.                  pr_agent/algo/utils.py Introduced ReasoningEffort enum with levels: HIGH, MEDIUM, LOW. +5/-0
Configuration changes	configuration.toml Extended configuration for reasoning effort setting.          pr_agent/settings/configuration.toml Added reasoning_effort configuration option with default value. +1/-0

---

Need help?

Type /help how to ... in the comments thread for any questions about Qodo Merge usage.

Check out the documentation for more information.

[qodo-merge-pro-for-open-source]

PR Reviewer Guide 🔍

Here are some key observations to aid the review process:

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

🧪 No relevant tests

🔒 No security concerns identified

⚡ Recommended focus areas for review Error Handling The reasoning_effort parameter is added without any error handling if the setting is invalid. Consider adding error handling for unsupported reasoning_effort values. if (model in self.support_reasoning_models): supported_reasoning_efforts = [ReasoningEffort.HIGH.value, ReasoningEffort.MEDIUM.value, ReasoningEffort.LOW.value] reasoning_effort = get_settings().config.reasoning_effort if (get_settings().config.reasoning_effort in supported_reasoning_efforts) else ReasoningEffort.MEDIUM.value get_logger().info(f"Add reasoning_effort with value {reasoning_effort} to model {model}.") kwargs["reasoning_effort"] = reasoning_effort

[qodo-merge-pro-for-open-source]

PR Code Suggestions ✨

Explore these optional code suggestions:

Category	Suggestion	Impact
General	Add error handling for unsupported models Add error handling for when reasoning_effort is used with an unsupported model. Currently, the code silently ignores reasoning_effort for unsupported models. pr_agent/algo/litellm_ai_handler.py [237-242] # Add reasoning_effort if model supports it -if (model in self.support_reasoning_models): - supported_reasoning_efforts = [ReasoningEffort.HIGH.value, ReasoningEffort.MEDIUM.value, ReasoningEffort.LOW.value] - reasoning_effort = get_settings().config.reasoning_effort if (get_settings().config.reasoning_effort in supported_reasoning_efforts) else ReasoningEffort.MEDIUM.value - get_logger().info(f"Add reasoning_effort with value {reasoning_effort} to model {model}.") - kwargs["reasoning_effort"] = reasoning_effort +if get_settings().config.reasoning_effort: + if model not in self.support_reasoning_models: + get_logger().warning(f"Model {model} does not support reasoning_effort. Ignoring setting.") + else: + supported_reasoning_efforts = [ReasoningEffort.HIGH.value, ReasoningEffort.MEDIUM.value, ReasoningEffort.LOW.value] + reasoning_effort = get_settings().config.reasoning_effort if (get_settings().config.reasoning_effort in supported_reasoning_efforts) else ReasoningEffort.MEDIUM.value + get_logger().info(f"Add reasoning_effort with value {reasoning_effort} to model {model}.") + kwargs["reasoning_effort"] = reasoning_effort [To ensure code accuracy, apply this suggestion manually] Suggestion importance[1-10]: 7 __ Why: The suggestion improves error handling by adding a warning log when reasoning_effort is configured for unsupported models, making the system behavior more transparent and helping with debugging. The improved code also adds a check for the reasoning_effort configuration existence.	Medium
Learned best practice	Add proper error handling and null checks when accessing configuration settings to prevent potential runtime errors Add error handling around the settings access to handle cases where config or reasoning_effort settings might be missing or invalid. Also validate the model parameter to ensure it's not None. pr_agent/algo/litellm_ai_handler.py [237-241] -if (model in self.support_reasoning_models): - supported_reasoning_efforts = [ReasoningEffort.HIGH.value, ReasoningEffort.MEDIUM.value, ReasoningEffort.LOW.value] - reasoning_effort = get_settings().config.reasoning_effort if (get_settings().config.reasoning_effort in supported_reasoning_efforts) else ReasoningEffort.MEDIUM.value - get_logger().info(f"Add reasoning_effort with value {reasoning_effort} to model {model}.") - kwargs["reasoning_effort"] = reasoning_effort +if model and model in self.support_reasoning_models: + try: + settings = get_settings() + supported_reasoning_efforts = [ReasoningEffort.HIGH.value, ReasoningEffort.MEDIUM.value, ReasoningEffort.LOW.value] + reasoning_effort = settings.config.reasoning_effort if ( + hasattr(settings, 'config') and + hasattr(settings.config, 'reasoning_effort') and + settings.config.reasoning_effort in supported_reasoning_efforts + ) else ReasoningEffort.MEDIUM.value + get_logger().info(f"Add reasoning_effort with value {reasoning_effort} to model {model}.") + kwargs["reasoning_effort"] = reasoning_effort + except Exception as e: + get_logger().warning(f"Error setting reasoning_effort, using default: {e}") + kwargs["reasoning_effort"] = ReasoningEffort.MEDIUM.value [To ensure code accuracy, apply this suggestion manually] Suggestion importance[1-10]: 6	Low
Update

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

[KennyDizi]

I checked and guess that pipelines have fallen because this PR #1529

[KennyDizi]

Through enum fields of Reasoning's delight,
Where HIGH and LOW efforts take their flight 🚀
In TOML's garden, configurations bloom,
While LiteLLM handlers weave their loom,
Such craftsmanship in each parameter true -
A masterwork of code, both wise and new! 💫 🎭

[KennyDizi]

Would you pls take a look at this PR @mrT23

[mrT23]

@KennyDizi
looks good

can you add a short new section here: https://qodo-merge-docs.qodo.ai/usage-guide/changing_a_model/
to document these dedicated parameters ? otherwise i think they will be less accessible

something like:

## Dedicated parameters

### OpenAI models
[config]
reasoning_efffort= ...

[KennyDizi]

@KennyDizi looks good

can you add a short new section here: https://qodo-merge-docs.qodo.ai/usage-guide/changing_a_model/ to document these dedicated parameters ? otherwise i think they will be less accessible

something like:

## Dedicated parameters

### OpenAI models
[config]
reasoning_efffort= ...

Thanks for your advice, @mrT23; I will update it soon.

[KennyDizi]

@mrT23 I've added a concise document for OpenAI models dedicated params.

[mrT23]

@KennyDizi
great.

Claude Sonnet also introduced some thinking mechanism (with a different API ...), but lets save it to a different PR
https://docs.litellm.ai/docs/providers/anthropic#usage---thinking--reasoning_content