Private AI
Kimi K2.5 with extended thinking capabilities. Moonshot AI's native multimodal model with strong general reasoning, visual coding, and agentic tool-calling. Running inside a TEE (Trusted Execution Environment), with provider attestation support.
Added Jan 29, 2026
Model weightsContext Window
128.0K
Max Output
65.5K
Avg output tokens (7d)
5.3K tokens
Input Price (Auto)
$0.60/1M
Output Price (Auto)
$3.00/1M
Cache Read (Auto)
$0.30/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
35.4
Auto routing is available for this model. Explicit provider selection is not available.
Loading provider options…
Coding Index
46.8
GPQA Diamond
Graduate-level scientific reasoning
87.9%
Better than 90% of models compared
HLE
Humanity's Last Exam
29.4%
Better than 90% of models compared
IFBench
Instruction-following benchmark
70.2%
Better than 84% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
95.9%
Better than 96% of models compared
AA-LCR
Long context reasoning evaluation
65.3%
Better than 83% of models compared
SciCode
Python programming for scientific computing
49.0%
Better than 91% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
34.8%
Last updated Jul 15, 2026
Artificial AnalysisBetter than 80% of models compared
