Re:Show HN: Route your prompts to the best LLM
by mzl on Thursday May 23, 2024 @03:59AM
(#40451852)
Sounds similar to the Composition of Experts from SambaNova (although that is for increased accuracy on single systems mostly, not for decreased cost from various suppliers): https://sambanova.ai/blog/samba-coe-the-power-of-routing-ml-...
[ Parent
| on HN ]
Re:Show HN: Route your prompts to the best LLM
by danlenton on Thursday May 23, 2024 @04:15AM
(#40451969)
Yep it's the same idea. We also see improvements in quality on some benchmarks beyond any specific model. This is common especially common when training a custom router on your own prompt dataset: https://youtu.be/9JYqNbIEac0