llm-merging

Model Merging is a cutting-edge technique that integrates multiple large language models (LLMs) into a single, cohesive model. This approach allows for the combination of models with different architectures, such as LLaMA 2, Mistral, and Wizard, to harness the strengths of each.

🤖 Self-hosted web UI for merging open-weight LLMs with zero GPU required, crash-safe pipelines, and 1-click safetensors export.