AI research has gone on at such a breakneck speed in recent years that it's difficult to keep up. Not a week seems to go by without some new benchmark-breaking LLM, so I thought it would be cool to trace the lineage of all of these models in one chart.
What makes one LLM a relative of another is up for debate, and likely I'm missing some connections or over-emphasizing others. No doubt this will look completely different in the years to come. If you have any comments or suggestions feel free to let me know, either here or on GitHub.
What makes one LLM a relative of another is up for debate, and likely I'm missing some connections or over-emphasizing others. No doubt this will look completely different in the years to come. If you have any comments or suggestions feel free to let me know, either here or on GitHub.