# Jinjie Ni | Knowledge Base | Instagit

Scaling Gemini at Google DeepMind.

---

## Repositories

### [megadlms](/jinjieni/megadlms)

GPU-optimized framework for training diffusion language models at any scale. The backend of Quokka, Super Data Learners, and OpenMoE 2 training.

- 22 articles | 327 stars
- Tags: deep-dive, architecture, performance, how-to-guide, tutorial