Kimi K2.6

Moonshot
VisionTool useReasoningPrompt cachingStructured output
Context
262K
Max output
262K
Serving providers
0
Cheapest input
n/a /1M

Benchmarks

independent evaluations · source: Artificial Analysis and Hugging Face leaderboards
Speed & latency
Output speed
46 t/s
tokens / second
Time to first token
2.79s
latency
Headline indices
61.8
of 100
30.3
of 100
Individual benchmarks · click a row for effort variants
BenchmarkBest scoreEffortCost / task
AA-Briefcase16%
AIME 202696%default
APEX Agents28%default
CritPt8%default
EvalComputeProxy8749.6default
GDPval1190.4default
GPQA Diamond91%default
Humanity's Last Exam36%default
IFBench76%default
IT-Bench SRE31%default
Long-Context Reasoning70%default
MMMU-Pro79%default
Omniscience6.4default
SciCode53%default
SWE-bench Pro59%default
SWE-bench Verified80%default
TerminalBench Hard44%default
TerminalBench v2.166%default
τ-bench Banking21%default
τ²-bench96%default

Pricing by serving provider

availability
No public price yet. Not offered on-demand at any tracked provider.

Price history

input + output $/1M since we started tracking
Input Output
$0.000$1.00$2.00$3.00$4.00Apr 22Jun 3Jun 10$4.00$0.950

Frequently asked questions

Kimi K2.6 pricing, context and availability

What is the context window of Kimi K2.6?

Kimi K2.6 accepts up to 262K tokens of context and can return up to 262K output tokens.

What can Kimi K2.6 do?

Kimi K2.6 supports image input (vision), tool use, reasoning, prompt caching and structured output.

Other Moonshot models

compare pricing across the Moonshot lineup
Pricing verified against LiteLLM + OpenRouter + provider pages · refreshed daily

Every weekday

AI moves fast. Here's your debrief.

News, analysis, tools, and more.

For people who build with AI