Back
LLM Optimization Research (Rakuten)
2025
PythonPytorchLLMResearchOptimization
Research project on LLM optimization techniques, conducted as part of final year project with Rakuten.
Conducted extensive research on various optimization techniques for Large Language Models (LLMs) in collaboration with Rakuten. The project involved analyzing and implementing different approaches to improve model performance, reduce computational requirements, and enhance inference speed.
Key Features
- Model quantization
- Knowledge distillation
- Pruning techniques
- Performance analysis
Technology Stack
AI/ML
PyTorchLLMCustom Optimization
Tools
GitJupyter Notebooks
Challenges
- Balancing model size and performance
- Implementing optimization techniques
- Evaluating trade-offs
Key Learnings
- Advanced LLM optimization
- Research methodology
- Performance analysis
Screenshots
You can swipe to move between images