# Best engine for Flint  for RTX3090? — retained record

> Retained reddit record from reddit:LocalLLaMA.

Canonical: https://highsignal.app/data/records/1af82d43a0fe6f00

[← sources](https://highsignal.app/data)


source: [reddit](https://highsignal.app/data/reddit) (reddit:LocalLLaMA)  published 2026-10-23 15:18 ingested 2026-10-24 02:40

[original source ↗](https://www.reddit.com/r/LocalLLaMA/comments/1xdgxf3/best_engine_for_flint_for_rtx3090/)

## retained text



```
This [post](https://www.reddit.com/r/LocalLLaMA/comments/1xc23iq/recursive_flint_the_qwen38_35b_weve_all_been/) & [Sam Witteveen](https://www.youtube.com/@samwitteveenai)'s [video](https://youtu.be/BsfjQDSeTwI) talked me into trying out the new Flint model... has anyone got it running faster than the typical vLLM / lamma.cpp engines?  Seems specialized inference engines  are all the rage...
```



document fetched 2026-10-24 10:43 · canonical https://reddit.com/r/LocalLLaMA/comments/1xdgxf3/best_engine_for_flint_for_rtx3090
