MiMo-v2.5-Pro-UltraSpeed: 1T model with 1000 tokens per second

·Hacker News··

MiMo, in collaboration with TileRT, releases the UltraSpeed mode of Xiaomi MiMo-V2.5-Pro — breaking 1000 tokens/s generation speed on a 1T-parameter model for the first time on commodity GPUs through extreme model-system codesign.

Read full article →

Related Articles

Twenty One Zero-Days in FFmpeg
redbell · Hacker News · 16h ago
CRISPR tech selectively shreds cancer cells, including "undruggable" cancers
gmays · Hacker News · 22h ago
Arch Linux Now Believes Malware Incident Under Control: More Than 1,500 Packages
qwertox · Hacker News · 2h ago
Kimi K2.7-Code: open-source coding model with better token efficiency
nekofneko · Hacker News · 1d ago
Swift at Apple: Migrating the TrueType hinting interpreter
DASD · Hacker News · 18h ago