MiniMax M2.5 is a productivity-focused flagship model with stronger coding and office workflow performance (Word, Excel, PowerPoint), plus improved tool planning and token efficiency. Running inside a TEE (Trusted Execution Environment), with provider attestation support.
Added Apr 20, 2026
Context Window
196.6K
Max Output
131.1K
Input Price
$0.20/1M
Output Price
$1.38/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
41.9
Coding Index
37.4
GPQA Diamond
Graduate-level scientific reasoning
84.8%
Better than 89% of models compared
HLE
Humanity's Last Exam
19.1%
Better than 85% of models compared
IFBench
Instruction-following benchmark
71.6%
Better than 90% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
95.3%
Better than 96% of models compared
AA-LCR
Long context reasoning evaluation
66.0%
Better than 89% of models compared
SciCode
Python programming for scientific computing
42.6%
Better than 88% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
34.8%
Better than 85% of models compared
Last updated May 11, 2026
Artificial Analysis