Cerebras-GPT
The Cerebras-GPT family of models was developed by the AI accelerator company Cerebras following Chinchilla scaling laws as a demonstration of its Wafter-Scale Cluster technology.
A side-by-side comparison of release history, capabilities, licensing and model sizes.
| Attribute |
 Cerebras-GPT
|
 Koala
|
|---|---|---|
| Developer | Cerebras | BAIR |
| Current generation | – | – |
| Initial release | 2023-03-28 | 2023-04-03 |
| Instruct tuned | No | Yes |
| Coding | No | No |
| Finetuning | Yes | Yes |
| Open source | Yes | Yes |
| License | Apache 2.0 | Noncommercial |
| Model sizes | 1.3B, 2.7B, 6.7B, 13B | 7B, 13B |
| Links |
The Cerebras-GPT family of models was developed by the AI accelerator company Cerebras following Chinchilla scaling laws as a demonstration of its Wafter-Scale Cluster technology.
Released alongside Vicuna, Koala is one of many descendants of the Meta LLaMA model trained on dialogue data collected from the web. On the developers' benchmarks, Koala outperforms its sibling Alpaca, though its adoption has been significantly less than that of its other sibling, Vicuna. Due to its use of LLaMA, only research use is permitted.