Requesty
DeepInfra Inc. logo
DeepInfra Inc.

moonshotai/Kimi-K2.6:flex

Kimi K2.6 is an open-source, native multimodal agentic model that advances practical capabilities in long-horizon coding, coding-driven design, proactive autonomous execution, and swarm-based task orchestration.

VisionReasoningTool callingCachingJSON schema
CompareDocs

Specifications

Context window262K tokens
Max outputN/A
API typechat
AddedAug 14, 2026
Model ID
Data retentionNo
Used for trainingNo
Provider location🇺🇸 US
Privacy policyDeepInfra Privacy Policy→

Benchmarks

Released 2026-04-20
Coding Indexcoding
61.8%

Artificial Analysis Coding Index — a composite of coding evaluations including LiveCodeBench, SciCode and Terminal-Bench.

GPQA Diamondreasoning
91.1%

Graduate-level physics, chemistry & biology questions designed to resist Googling.

Intelligence Indexreasoning
45.1%

Artificial Analysis Intelligence Index — a composite of multiple evaluations measuring overall model capability.

Scores are sourced from official model cards, Artificial Analysis, and public leaderboards. Benchmarks measure specific skills and do not capture every aspect of model quality. Always test on your own workload.

Pricing

Prices updated August 15, 2026
Input / 1M
$0.60
Output / 1M
$2.80
Cache write
N/A
Cache read / 1M
$0.12
Estimated cost
100K input + 10K output$0.0880
1M input + 100K output$0.88
10M input + 1M output$8.80

Requesty charges exactly what the upstream provider charges, no markup, no per-request fees. Prompt caching and smart routing can reduce effective cost by 30-80%.

Quickstart

Drop-in compatible with the OpenAI SDK. Change the base URL, swap in your Requesty API key, and set the model to deepinfra/moonshotai/Kimi-K2.6:flex.

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_REQUESTY_API_KEY",
    base_url="https://router.requesty.ai/v1",
)

response = client.chat.completions.create(
    model="deepinfra/moonshotai/Kimi-K2.6:flex",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

print(response.choices[0].message.content)

Other DeepInfra Inc. models

Qwen/Qwen3.5-397B-A17B:flex
262K ctx$0.36/1M
Coding 48
XiaomiMiMo/MiMo-V2.5-Pro:flex
1.0M ctx$0.80/1M
Coding 60
deepseek-v4-flash-0731:flex
1.0M ctx$0.07/1M
Coding 69
Qwen/Qwen3-Coder-480B-A35B-Instruct-Turbo:flex
262K ctx$0.24/1M
deepseek-ai/DeepSeek-V4-Flash:flex
1.0M ctx$0.07/1M
Coding 69
meta-llama/Meta-Llama-3.1-8B-Instruct-Turbo:flex
131K ctx$0.02/1M

Frequently asked questions

How much does moonshotai/Kimi-K2.6:flex cost?
moonshotai/Kimi-K2.6:flex is priced at $0.60 per million input tokens and $2.80 per million output tokens when accessed via Requesty. Prompt caching is supported, which can cut effective input cost by up to 90% on repeated context. Requesty charges exactly what the upstream provider charges, we don't add markup.
What is the context window of moonshotai/Kimi-K2.6:flex?
moonshotai/Kimi-K2.6:flex has a context window of 262K tokens. That's roughly 350 words of input you can fit in a single prompt.
How does moonshotai/Kimi-K2.6:flex perform on benchmarks?
moonshotai/Kimi-K2.6:flex scores 95.9% on τ²-Bench, 91.1% on GPQA Diamond, 61.8% on Coding Index. See the full benchmark chart above for results across MMLU Pro, GPQA Diamond, SWE-Bench Verified, HumanEval, MATH, AIME, MMMU, and LiveBench.
What can moonshotai/Kimi-K2.6:flex do?
moonshotai/Kimi-K2.6:flex supports vision input, tool calling, extended reasoning, prompt caching, structured outputs (JSON schema). You can call it through any OpenAI-compatible client by pointing base_url to Requesty.
How do I use moonshotai/Kimi-K2.6:flex with the OpenAI SDK?
Install the OpenAI SDK, set base_url to "https://router.requesty.ai/v1", set your API key to your Requesty key, and set the model to "deepinfra/moonshotai/Kimi-K2.6:flex". The Quickstart above shows Python, JavaScript and cURL snippets.
Can I run moonshotai/Kimi-K2.6:flex through Requesty?
Yes. moonshotai/Kimi-K2.6:flex runs through Requesty's OpenAI-compatible API, served from DeepInfra Inc.. You do not host the model yourself: point base_url at Requesty, set the model to "deepinfra/moonshotai/Kimi-K2.6:flex", and requests are routed to the upstream provider with automatic failover. The same key gives you 600+ other models too.

Access moonshotai/Kimi-K2.6:flex through Requesty

One API key, 600+ models, OpenAI-compatible. No markup on provider prices, automatic failover, and smart caching built-in.

All DeepInfra Inc. models