Kimi K2.5
Kimi · 2 thinking configurations
Intelligence23.5
Input / 1M$0.60
Output / 1M$2.75
Context256K
Tariff: Kimi · Source: artificial-analysis · USD per million tokens
Thinking configuration
What does more thinking buy you?
Compare this model’s measured thinking configurations.
Scroll the chart ↔ · tap a point for details
Efficient frontier: best measured score at each cost2 measured configurations · 0 without both values
Selected: thinking thinking · Click a point to switch configuration
Kimi K2.5 is Moonshot AI's native multimodal model, delivering state-of-the-art visual coding capability and a self-directed agent swarm paradigm. Built on Kimi K2 with continued pretraining over approximately 15T mixed...
Intelligence IndexArtificial Analysis composite index, not a percentage. Different versions of the index are not directly comparable.
23.5 pointsGPQA DiamondGraduate-level scientific questions. Measures specialist reasoning rather than web-search quality.
87.9 %Humanity’s Last ExamDifficult questions across many disciplines, compared under the same evaluation setup.
30.7 %Terminal-Bench 2.1Agent tasks in a terminal environment. Results also depend on the evaluation harness.
45.7 %τ²-BenchTool use and interaction in assistance scenarios. Does not replace an evaluation of your integration.
95.9 %IFBenchVerifiable instruction following. A proxy for precision rather than a benchmark of creative style.
70.2 %MMMU-ProMultimodal understanding of academic problems that include images.
75.4 %Omniscience accuracyFactual accuracy in Artificial Analysis's Omniscience evaluation.
35.2 %Specifications from the catalog
Inputtext, image
Outputtext
Tool callingYes
Structured outputYes
SpeedNot available
First tokenNot available
Maximum output235.9K
Catalog listing date2026-01-27