FLAN-UL2
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.
A side-by-side comparison of release history, capabilities, licensing and model sizes.
| Attribute |
 FLAN-UL2
|
 GPT-J
|
|---|---|---|
| Developer | EleutherAI | |
| Current generation | – | – |
| Initial release | 2023-03-03 | 2021-06-09 |
| Instruct tuned | Yes | Yes |
| Coding | No | No |
| Finetuning | Yes | Yes |
| Open source | Yes | Yes |
| License | Apache 2.0 | Apache 2.0 |
| Model sizes | 20B | 6B |
| Links |
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.
GPT-J is a model released by EleutherAI shortly after its release of GPTNeo, with the aim of delveoping an open source model with capabilities similar to OpenAI's GPT-3 model. With a larger size than GPTNeo, GPT-J also performs better on various benchmarks.