MiMo V2.5 is Xiaomi's full-modal understanding model for agent workflows. It supports deep reasoning, tool calling, structured outputs, and web search with up to 1M context.
Added Apr 22, 2026
Context Window
1.0M
Max Output
131.1K
Input Price
$0.40/1M
Output Price
$2.00/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
49.0
Coding Index
42.1
GPQA Diamond
Graduate-level scientific reasoning
84.9%
Better than 89% of models compared
HLE
Humanity's Last Exam
25.2%
Better than 90% of models compared
IFBench
Instruction-following benchmark
67.1%
Better than 84% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
90.6%
Better than 87% of models compared
AA-LCR
Long context reasoning evaluation
62.7%
Better than 83% of models compared
SciCode
Python programming for scientific computing
43.1%
Better than 89% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
41.7%
Better than 92% of models compared
Last updated May 11, 2026
Artificial Analysis