[GUEST ACCESS MODE: Data is scrambled or limited to provide examples. Make requests using your API key to unlock full data. Check https://lunarcrush.ai/auth for authentication information.]

![slow_developer Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::1461305890892570628.png) Haider. [@slow_developer](/creator/twitter/slow_developer) on x 46.7K followers
Created: 2025-07-22 15:55:07 UTC

Google Logan Paul says model progress now comes from scaling all three stages:

"pre-training, post-training, and reasoning with RL"

Gemini XXX succeeded by pushing all three in parallel.

post-training gave good results, but reasoning is now easier to scale

that's why companies are investing in all three, because they all work

![](https://pbs.twimg.com/amplify_video_thumb/1947685842069196802/img/XYK9TB-phk9vVHdD.jpg)

XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1947686856809701503/c:line.svg)

**Related Topics**
[investment](/topic/investment)
[logan paul](/topic/logan-paul)
[$googl](/topic/$googl)
[stocks communication services](/topic/stocks-communication-services)

[Post Link](https://x.com/slow_developer/status/1947686856809701503)