# Together AI says to test models on your own workloads, not leaderboards alone

> Original title: public benchmarks are useful for building a model shortlist

- Source: Together AI
- Published: 2026-11-09T04:00:37.000Z
- Bad Influence: https://badinfluence.news/items/ckpnatjtn53yy8o4p7k27nai2
- Original: https://x.com/togethercompute/status/2119641283786362909

## Summary

Together AI says public benchmarks help build a model shortlist, but the real test is replaying the workloads you serve and checking which models pass your own evals. It advises using leaderboards to shortlist and live traffic to decide what to ship.

## Post

public benchmarks are useful for building a model shortlist

the real test is replaying the workloads you actually serve and seeing which models pass your own evals

leaderboards help you shortlist. your traffic tells you what to ship

![](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Fmedia%2FHWp6g71WMAA5v_6.jpg&mode=thumb&exp=1794528000&sig=a197bdea4165f7b1)