FLAN-UL2 vs. Guanaco

Overview

Similar to FLAN-T5, FLAN-UL2 is a model based on Google's popular T5 architecture with an upgraded pre-training procedure dubbed UL2. On most NLU benchmarks, FLAN-UL2 outperforms FLAN-T5 by a significant margin.

Initial release: 2023-03-03

Reference

https://huggingface.co/google/flan-ul2

Further Reading

• https://www.yitay.net/blog/flan-ul2-20b

Overview

Guanaco is an LLM based off the QLoRA 4-bit finetuning method developed by Tim Dettmers et. al. in the UW NLP group. Guanaco achieves 99% ChatGPT performance on the Vicuna benchmark.

Guanaco is an LLM that uses a finetuning method called LoRA that was developed by Tim Dettmers et. al. in the UW NLP group. With QLoRA, it becomes possible to finetune up to a 65B parameter model on a 48GB GPU without loss of performance relative to a 16-bit model. The Guanaco model family outperforms all previously released models on the Vicuna benchmark. However, given the models are based off of the LLaMA model family, commercial use is not permitted.

Initial release: 2023-05-23

Reference

https://github.com/artidoro/qlora

Further Reading

• https://arxiv.org/abs/2305.14314
• https://twitter.com/Tim_Dettmers/status/1661379354507476994