kimi/kimi-k2.6

kimi/kimi-k2.6
FlagshipFeatured
VisionToolsJSONReasoning
by kimi

Moonshot Kimi K2 Thinking — most advanced open reasoning model in the K2 series, agentic long-horizon tasks, 256k context.

ctx262.1K tokens
Max output32.8K
Inputtext + image + video
Outputtext
p50 TTFT2.91 s
INPUT$0.95/ 1M tokens
OUTPUT$4.00/ 1M tokens
p50 TTFT2.91 s7d
p95 TTFT10.00 s7d
TRAFFIC4.8Mtokens / 7d

Kimi K2.6 is a flagship multimodal model from Kimi, optimized for tasks that require reasoning over long contexts and multiple input modalities. It processes text, images, and video, with a context…

What is Kimi K2.6 and who is it designed for?

What input modalities does Kimi K2.6 support?

How does Kimi K2.6 compare to other flagship models on OrcaRouter?

Why would a developer choose Kimi K2.6 over smaller or cheaper models?

Code samples

Call from any SDK

OpenAI-compatible — keep the SDK you already use

  • OpenAI SDKhttps://api.orcarouter.ai/v1
from openai import OpenAI

client = OpenAI(
    base_url="https://api.orcarouter.ai/v1",
    api_key="$ORCAROUTER_API_KEY",
)

response = client.chat.completions.create(
    model="kimi/kimi-k2.6",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Supported parameters

  • frequency_penalty
  • include_reasoning
  • max_completion_tokens
  • max_tokens
  • n
  • presence_penalty
  • prompt_cache_key
  • reasoning
  • response_format
  • safety_identifier
  • stop
  • stream
  • stream_options
  • temperature
  • thinking
  • tools
  • top_p

Pricing

Input / 1M tokens$0.950
Output / 1M tokens$4.00
Cache read / 1M$0.160
CurrencyUSD

Cost calculator

Tokens / month10MM
Input share70%%
Estimated / month $18.65 · With prompt caching $15.89

Estimate based on list price

Token & cost estimator

Input tokens: 20Cost per request: $0.002019

Estimate only — actual token counts depend on the provider's tokenizer.

Performance

p50 TTFT
2.91 s
Output speed
42.9 tok/s
p95 TTFT
10.00 s
Error rate
0.77%

Public benchmarks

47.1
AA Coding
Better than 62% of models compared
#40 of 106
53.9
AA Intelligence
Better than 73% of models compared
#30 of 110
GPQA Diamond
91.1
Humanity's Last Exam
35.9
IFBench
76.0
Long-Context Recall
69.7
SciCode
53.5
TerminalBench Hard
43.9
τ²-Bench
95.9
Source: artificialanalysis.ai

How it compares

kimi/kimi-k2.6Kimi K3Kimi K2.7 Codekimi/kimi-k2.5
Input $/M$0.95$3.00$0.95$0.60
Output $/M$4.00$15.00$4.00$3.00
Context262K1.0M262K262K
Quality8/109/108/107/10
Compare side-by-sideCompare side-by-sideCompare side-by-sideCompare side-by-side

FAQ

What is the cost of using Kimi K2.6 on OrcaRouter?
Kimi K2.6 costs $0.95 per 1 million input tokens and $4.00 per 1 million output tokens, billed at the provider rate with zero markup through OrcaRouter. There are no additional fees.
What is the context window size and maximum output tokens for Kimi K2.6?
The context window is 262,144 tokens, and the maximum output is 32,768 tokens.
What are the main strengths of Kimi K2.6?
Its main strengths are a large context window (262k tokens), strong tool-use reasoning (τ²-Bench score 95.9), and support for text, image, and video inputs.
How does Kimi K2.6 compare to other flagship models like GPT-4o or Claude?
Kimi K2.6 has a larger context window than GPT-4o (128k) and Claude Opus (200k) and supports video input. Its pricing is lower than those models in many cases, but direct benchmark comparisons beyond τ²-Bench are not available.
What data handling practices apply when using Kimi K2.6 through OrcaRouter?
OrcaRouter forwards your request to the Kimi provider and does not store prompts or responses beyond necessary transmission. Review Kimi's privacy policy for their data handling. OrcaRouter does not train on your data.
How do I call Kimi K2.6 via an OpenAI-compatible API?
Use base URL https://api.orcarouter.ai/v1, model ID "kimi/kimi-k2.6", and an OrcaRouter API key. The API is fully OpenAI-compatible; you can use the same message formats and parameters.
Does OrcaRouter offer any discounts or caching for Kimi K2.6?
No, there are no discounts, caching, or volume pricing for this model. You pay per token at the listed rates with zero markup.
Can Kimi K2.6 process audio or voice inputs?
No, Kimi K2.6 only accepts text, image, and video inputs. Audio must be transcribed to text before being passed to the model.
What is the latency of Kimi K2.6?
Exact latency figures are not published. It is a flagship model, so response times are higher than smaller models. Expect longer times for large context or multimodal inputs.
Is Kimi K2.6 suitable for real-time applications?
Due to potentially high latency and token costs, it is not ideal for real-time or high-frequency applications. It is better suited for offline batch processing or tasks where accuracy is prioritized over speed.

Embed this badge

kimi/kimi-k2.6$0.95/M in2906ms p50via OrcaRouter
HTML <a href="https://www.orcarouter.ai/models/kimi/kimi-k2.6" target="_blank"> <img src="https://www.orcarouter.ai/embed/kimi/kimi-k2.6.svg" alt="kimi/kimi-k2.6 on OrcaRouter" /> </a>
Markdown [![kimi/kimi-k2.6](https://www.orcarouter.ai/embed/kimi/kimi-k2.6.svg)](https://www.orcarouter.ai/models/kimi/kimi-k2.6)

Model card as data

GET /api/public/models/kimi/kimi-k2.6Open
Machine-readable:/llms.txt/llms-full.txt