algorithmicsuperintelligence/optillm
Optimizes LLM reasoning without model training
View on index · View in 3D Map
Related repos
- Autonomous algorithm discovery with LLMs
- tools for building and evaluating advanced LLM reasoning algorithms
- OpenAI-compatible proxy that boosts LLM reasoning accuracy with 20+ optimization techniques, no retraining needed
- Collection of papers on LLM reasoning
- Builds reasoning LLM in PyTorch from scratch
- Forces LLMs into structured reasoning through argumentation models
- Environments for reinforcement learning with LLMs
- modular AI programming with LLMs, not just prompting
- zkML framework for proving AI model inferences with zero-knowledge proofs
- Automated framework for optimizing LLM model harnesses
