Kimi K2.6 Thinking is the reasoning-optimized K2.6 variant for deeper multi-step planning and execution. It is tuned for long-horizon coding and design workflows, including complex orchestration across many specialized sub-agents and autonomous end-to-end output generation.
Added Apr 16, 2026
Context Window
256.0K
Max Output
65.5K
Input Price (Auto)
$0.53/1M
Output Price (Auto)
$2.73/1M
Capabilities
Performance metrics and benchmarks
Sourced from Artificial Analysis.
Intelligence Index
53.9
Choose explicit providers for this model. Auto routing remains available as the default option.
Loading provider options…
Coding Index
47.1
Agentic Index
66.0
GPQA Diamond
Graduate-level scientific reasoning
91.1%
Better than 98% of models compared
HLE
Humanity's Last Exam
35.9%
Better than 98% of models compared
IFBench
Instruction-following benchmark
76.0%
Better than 96% of models compared
T²-Bench Telecom
Conversational AI agents in dual-control scenarios
95.9%
Better than 97% of models compared
AA-LCR
Long context reasoning evaluation
69.7%
Better than 95% of models compared
GDPval-AA
Economically valuable tasks
49.2%
Better than 94% of models compared
CritPt
Research-level physics reasoning
8.0%
Better than 95% of models compared
SciCode
Python programming for scientific computing
53.5%
Better than 98% of models compared
Terminal-Bench Hard
Agentic coding and terminal use
43.9%
Better than 94% of models compared
AA-Omniscience Accuracy
Proportion of correctly answered questions
32.8%
Better than 89% of models compared
AA-Omniscience Hallucination Rate
Rate of incorrect answers among non-correct responses
39.3%
Better than 94% of models compared
Last updated May 11, 2026
Artificial Analysis