# Lewis Tunstall backs supervised fine-tuning over expensive reinforcement learning

> Original title: As the resident “SFT all the things” person at HF, I approve of this message.

- Source: Lewis Tunstall
- Published: 2026-11-05T22:40:39.000Z
- Bad Influence: https://badinfluence.news/items/n341fxmfx8c5opawj83kgar1q
- Original: https://x.com/_lewtun/status/2118473597832327267

## Summary

Lewis Tunstall of Hugging Face says he approves of the message that supervised fine-tuning, or SFT, should be used for everything. A quoted post from Esong Ghori advises climbing as far as possible with SFT before relying on expensive reinforcement learning, or RL. Tunstall adds that RL is the cherry on top, citing Yann LeCun.

## Post

As the resident “SFT all the things” person at HF, I approve of this message.

## Quoted post @esonghori2

> One lesson from this launch: climb as far as you can with SFT before relying on expensive RL.
> 
> RL is still the cherry on top, as @ylecun once put it.

https://x.com/esonghori2/status/2118058972225908825