Peano Labs(@peano_ai):We enable full-parameter RL on TPUs: MiMo-V2.6 at 310B, plus other stable training runs of 1,000+ steps across 1,000+ TPUs. With JAX, scaling up is a config change, not a rewrite. We built on that with optimized vLLM inference for faster rollouts and full bitwise trainer–sampler agreement in validation. Trainer and sampler share one TPU ICI fabric. All 310B MiMo-V2.6 parameters transfer in <2 seconds.

2026.09.21 21:28

We enable full-parameter RL on TPUs: MiMo-V2.6 at 310B, plus other stable training runs of 1,000+ steps across 1,000+ TPUs. With JAX, scaling up is a config change, not a rewrite. We built on that with optimized vLLM inference for faster rollouts and full bitwise trainer–sampler agreement in validation. Trainer and sampler share one TPU ICI fabric. All 310B MiMo-V2.6 parameters transfer in <2 seconds.