MiMo-V2.6: Scaling Reinforcement Learning Towards Self-Improvement

English summary

Rare cost disclosure: RL post-training ~USD 2.6M (Pro, 1.02T/42B active) and about 0.9M (Flash); per step 1,568 prompts x16 rollouts, 2.7-3.7B tokens; rollout 43.8% vs training 43.5% of compute; 68% of RL batch is code tasks. Lab self-report kept for cost data. Found via InfoQ gNdWoDM8ygkAjXfsLaBK.