# Together AI argues token efficiency should top model evaluation criteria

> Original title: If you’re comparing models by their price sheet, you’re missing half the math.

- Source: Together AI
- Published: 2026-11-09T21:09:22.000Z
- Bad Influence: https://badinfluence.news/items/zx4w9z5k0yrnbmqc8bphvcwi3
- Original: https://x.com/togethercompute/status/2119900177565942018

## Summary

Together AI says comparing models by price sheet misses half the cost math, and that token efficiency belongs at the top of evaluation criteria. NVIDIA's Jacob Loewenstein made the case at Open by Design NYC, a discussion that also included Reflection's Zach Wentz and Together AI's Hassan El Mghari.

## Post

If you’re comparing models by their price sheet, you’re missing half the math.

Jacob Loewenstein @spatialjlo from NVIDIA explains why token efficiency belongs at the top of your evaluation criteria, whether you’re adopting a model or training your own.

One of the standout moments from Open by Design NYC, where Jacob, Zach Wentz @zkwentz from Reflection, and Hassan El Mghari @nutlope from Together AI chatted with builders digging into open models, post-training, and inference.

[Video](https://badinfluence.news/api/img-proxy?u=https%3A%2F%2Fpbs.twimg.com%2Famplify_video_thumb%2F2119899608021446656%2Fimg%2FHAfZTVwTnwv8X44f.jpg&mode=thumb&exp=1794528000&sig=611805540919a016)