Abhishek/server side cost calculation

[Mandolaro]

📝 Summary

• 1. Rewrite Cost Calculation

✅ Checklist

• Docs updated (if necessary)
• Changelogs are updated (if necessary)

[gemini-code-assist]

Summary of Changes

Hello @Mandolaro, I'm Gemini Code Assist¹! I'm currently reviewing this pull request and will post my feedback shortly. In the meantime, here's a summary to help you and other reviewers quickly get up to speed!

This pull request undertakes a substantial architectural refactoring aimed at enhancing the server-side cost calculation and LLM tracing framework. By modularizing the integration logic for various LLM providers into separate, dedicated sub-packages, the system gains improved maintainability and scalability. The changes also introduce more detailed token usage metrics, particularly for caching, which will enable a more precise understanding of resource consumption and cost implications.

Highlights

• Modular LLM Tracing Architecture: The core LLM tracing and cost calculation logic has been significantly refactored. Provider-specific implementations for OpenAI, Anthropic, Together, Google GenAI, and Groq are now moved into dedicated sub-packages, improving code organization, maintainability, and extensibility.
• Enhanced Usage Metrics: New attributes have been introduced to track cache-related token counts (e.g., GEN_AI_USAGE_CACHE_CREATION_INPUT_TOKENS, GEN_AI_USAGE_CACHE_READ_INPUT_TOKENS), providing more granular insights into LLM usage and potential cost optimizations.
• Decorator Module Refactoring: Utility decorators like use_once and dont_throw have been moved into their own distinct files within the judgeval.utils.decorators package, promoting a cleaner and more focused module structure.

Using Gemini Code Assist

The full guide for Gemini Code Assist can be found on our documentation page, here are some quick tips.

Invoking Gemini

You can request assistance from Gemini at any point by creating a comment using either /gemini <command> or @gemini-code-assist <command>. Below is a summary of the supported commands on the current page.

Feature	Command	Description
Code Review	/gemini review	Performs a code review for the current pull request in its current state.
Pull Request Summary	/gemini summary	Provides a summary of the current pull request in its current state.
Comment	@gemini-code-assist	Responds in comments when explicitly tagged, both in pull request comments and review comments.
Help	/gemini help	Displays a list of available commands.

Customization

To customize Gemini Code Assist for GitHub experience, repository maintainers can create a configuration file and/or provide a custom code review style guide (such as PEP-8 for Python) by creating and adding files to a .gemini/ folder in the base of the repository. Detailed instructions can be found here.

Limitations & Feedback

Gemini Code Assist may make mistakes. Please leave feedback on any instances where its feedback is incorrect or counter productive. You can react with 👍 and 👎 on @gemini-code-assist comments. If you're interested in giving your feedback about your experience with Gemini Code Assist for Github and other Google products, sign up here.

You can also get AI-powered code generation, chat, as well as code reviews directly in the IDE at no cost with the Gemini Code Assist IDE Extension.

Footnotes

1. Review the Privacy Notices, Generative AI Prohibited Use Policy, Terms of Service, and learn how to configure Gemini Code Assist in GitHub here. Gemini can make mistakes, so double check it and use code with caution. ↩

[gemini-code-assist]

Code Review

This pull request introduces a significant and well-structured refactoring to move LLM cost calculation to the server side. The client-side logic is now modularized into provider-specific wrappers, which greatly improves maintainability. The changes correctly focus on sending token usage data to the backend instead of calculating costs on the client.

I've identified a few issues:

• A critical issue where an e2e test will fail due to the removal of a cost-related attribute key.
• A high-severity issue regarding duplicated code in the new llm module structure.
• A high-severity regression where cached token tracking for OpenAI has been lost.
• A medium-severity issue regarding a removed type hint.

Overall, this is a great refactoring. Addressing these points will make it even better.

[abhishekg999]

lets see it in staging