feat(llm-enhancements): add intelligent prompt optimization to reduce token usage by up to 77%#3
Open
berkdurmus wants to merge 1 commit intofeat/enhancements-2from
Open
Conversation
… token usage by up to 77%
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
This PR introduces intelligent prompt optimization techniques that significantly reduce token usage while maintaining response quality. Our testing shows up to 77% reduction in prompt size for large traces, which translates to substantial cost savings and improved response times.
Key Changes
Created utility functions for text compression and smart truncation
Implemented a base PromptOptimizer class that intelligently formats trace data
Added specialized optimizers for different agent types:
DiagnosisPromptOptimizer
RecommendationPromptOptimizer
Updated agent classes to use the optimizers
Added comprehensive documentation
Created test script to demonstrate impact
Performance Improvements
Regular traces: 2-5% reduction in prompt size
Large traces: 75-80% reduction in prompt size
System prompts: 5-6% reduction in size
Optimization Techniques
Smart prioritization of critical information (errors, failed actions)
Content-aware compression that preserves diagnostic context
Intelligent truncation of repetitive data
System prompt optimization for conciseness