[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.]

![rohanpaul_ai Avatar](https://lunarcrush.com/gi/w:24/cr:twitter::2588345408.png) Rohan Paul [@rohanpaul_ai](/creator/twitter/rohanpaul_ai) on x 73.4K followers
Created: 2025-07-09 04:10:25 UTC

📊 Main Results

Trained only on 32K‑token documents and an 8K window, the 14B model keeps accuracy above XX% until 448K tokens and still sits around XX% at 3.5M tokens.

Competing long‑context baselines collapse well before 1M tokens even though their advertised windows reach that size.

![](https://pbs.twimg.com/media/GvY0oNPXkAAu61u.png)

XXX engagements

![Engagements Line Chart](https://lunarcrush.com/gi/w:600/p:tweet::1942798472362783006/c:line.svg)

**Related Topics**
[1m](/topic/1m)

[Post Link](https://x.com/rohanpaul_ai/status/1942798472362783006)