yuchenlin/LLM-Blender
An ensembling framework that combines multiple open-source LLMs using pairwise ranking and generative fusion to improve overall performance.
Not currently ranked — collecting fresh signals.
star history
LLM-Blender is a research framework for ensembling multiple open-source LLMs to achieve consistently superior performance. It uses PairRM, a pairwise reward model, to rank outputs from different LLMs and identifies their respective strengths and weaknesses. A generative fusion module then combines the best aspects of each model to enhance capability. Published at ACL 2023 by researchers from AI2-Mosaic and USC INK lab, with models and datasets available on Hugging Face.
Frequently asked
- What is yuchenlin/LLM-Blender?
- An ensembling framework that combines multiple open-source LLMs using pairwise ranking and generative fusion to improve overall performance.
- Is LLM-Blender open source?
- Yes — yuchenlin/LLM-Blender is open source, released under the Apache-2.0 license.
- What language is LLM-Blender written in?
- yuchenlin/LLM-Blender is primarily written in Python.
- How popular is LLM-Blender?
- yuchenlin/LLM-Blender has 990 stars on GitHub.
- Where can I find LLM-Blender?
- yuchenlin/LLM-Blender is on GitHub at https://github.com/yuchenlin/LLM-Blender.