# Xiaomi Is Streaming the Reinforcement Learning Run of Its Next Model in Public

The MiMo team says each training step pulls 1,568 prompts, samples 16 rollouts per prompt and processes roughly 2 billion generated tokens, with telemetry on a live dashboard.

- Published: 2026-09-18T06:27:34.772Z
- Canonical: https://polylog.news/ai/2026-09-18/xiaomi-is-streaming-the-reinforcement-learning-run-of-its-ne
- Publisher: Polylog (AI desk)
- Section: tech
- Sources: [Polylog editors](https://polylog.news), [Xiaomi MiMo live dashboard](https://mimo.xiaomi.com/rl/), [Fuli Luo (X)](https://x.com/_LuoFuli/status/2100296686719610932)

Xiaomi's MiMo team ended roughly half a year of public silence by putting the reinforcement learning post-training of its next model on a live dashboard. The run is in progress, and the telemetry includes reward curves updating in real time…

This story is for subscribers. Read it in full at https://polylog.news/ai/2026-09-18/xiaomi-is-streaming-the-reinforcement-learning-run-of-its-ne (subscription information: https://polylog.news/pricing).