Xiaomi MiMo-V2.5
Open multimodal agent model with 1M context.
A multimodal MiMo-V2.5 model series described as supporting text, image, video, and audio, with up to a 1M-token context window.
Pricing
Model profile · Current snapshot
Input / 1M
$0.14
Output / 1M
$0.28
Blended / 1M
$0.175
Output TPS
53.59
TTFT (s)
2.49
Model Intelligence
Context window
1,000,000 tokens
Arena ranking
30
Benchmarkable
No
Model level
family
Intelligence Index
29.5
Coding Index
56.8
GPQA
0.85
HLE
0.27
SciCode
0.44
IFBench
0.67
LCR
0.73
TerminalBench Hard
0.42
TAU2
0.91
Recent stories
1 linked story