MiMo V2.6 Flash

Changes from previous version

A smaller MoE than MiMo V2.6 Pro (309B/15B versus 1.02T/42B) at about a third of the token price, for high-volume agent loops that can give up a few points on Xiaomi's coding table.

Release Summary

Efficiency checkpoint of MiMo V2.6 Pro: 309B total / 15B active, the same omnimodal inputs and 1M-token context, MIT weights as MiMo-V2.6-Flash-RL. Xiaomi scores Flash at 67.9 on DeepSWE v1.1, 52.3 on AutomationBench v1.0.6, 28.8 on Terminal-Bench 4.0, and 87.6 on Terminal-Bench 2.1. Overseas API ID mimo-v2.6-flash is $0.14/M uncached input, $0.0028/M cache hits, and $0.28/M output. Batch inference is half of those uncached rates.

Timeline

Other MiMo versions