llamaperf

Mimo 2.5

Xiaomi · 1 report

Thin page (1 of 3 reports needed for indexing). Add yours.
Tone: positive
agenticcodinglong-context

Mimo 2.5 uses 5-to-1 sliding-window attention similar to Gemma 3, stays fast at large context on dual RTX Pro 6000. Also mentions Step 3.7 Flash (3-to-1) at ~40 t/s at 178k context. MiniMax M3 and DeepSeek V4 have kernel issues on Blackwell consumer GPUs.