kimi-k3

Name: kimi-k3
Brand: Nebius AI
SKU: nebius/kimi-k3
Price: 3.0000 USD
Availability: InStock

Nebius AI/🇪🇺 EU/chat

Kimi K3 is Moonshot AI's flagship reasoning model with a 1M token context window, strong agentic tool use, and long horizon task execution. Hosted on Nebius.More

Compare 7 providers All models →

Which id to call

nebius/kimi-k3

This exact deployment on Nebius AI, with no routing and no failover. Send it as the model field.

Input /1M

$3.00

Nebius AI

Output /1M

$15.00

5.0x input

Context

1.0M

131K output

Paid /1M

$3.24

measured, cache included

Capabilities 2/8

VisionReasoningTool callingCachingWeb searchJSON schemaComputer useImage generation

Provider rates

What kimi-k3 costs

Provider prices per 1M tokens, updated October 29, 2026.more

These are the upstream provider rates. Pay as you go adds 5%, or 0% if you bring your own keys, and there is no per-request fee. Prompt caching and routing change what you pay against these rates, not the rates themselves.

Input /1M

$3.00

Output /1M

$15.00

Cache write /1M

Cache read /1M

What a workload costs

100K input + 10K output: $0.45
1M input + 100K output: $4.50
10M input + 1M output: $45.00

At the rates above. This endpoint does not offer prompt caching, so repeated context is billed as new input every time.

model=

Which id to call

One base url, https://router.requesty.ai/v1, and one key for every id here.

nebius/kimi-k3

This exact deployment on Nebius AI, with no routing and no failover. Send it as the model field.

OpenAI compatible

Call it in three lines

Change the base url, use your Requesty key, set the model to any id on the left. Existing OpenAI SDK code needs no other edit.

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="nebius/kimi-k3",
    messages=[
        {"role": "user", "content": "Explain quantum computing in one paragraph."},
    ],
)

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

Live from production

Nebius AI on kimi-k3, measured

What this provider was measured doing on kimi-k3 across Requesty traffic.more

Whole-window figures, because that is the grain published per provider. They cover Nebius AI serving this model in every region it serves it from, so a region-pinned deployment shares them with its siblings. A figure is absent where no qualifying sample exists.

First token

1.27s

median

p95 wait

53.55s

slowest 5%

Output speed

84/s

median tokens

Paid /1M

$3.24

blended, cache included

Cache hit

90.1%

of input tokens

Compare these against the other 6 providers serving kimi-k3.

Reference

Specs and data terms

What the catalog reports for this deployment, and what the provider does with the traffic.

Context window1.0M tokens

Max output131K tokens

API typechat

AddedJul 2026

Model id

Hugging Facemoonshotai/Kimi-K3 →

Data retentionNone

Used for trainingNo

Served from🇪🇺 EU

Privacy policyNebius Privacy Policy →

Public leaderboards

Benchmark scores

Benchmarks measure the model, not this endpoint, so they are the same wherever these weights are served.

No benchmarks are published for this exact variant yet.

Region-specific deployments and highspeed tiers usually share scores with their base model. Try the base model page or the Nebius AI models overview.

Same provider

More from Nebius AI

Newest first, on the same provider and the same key.

glm-5.2

1M ctx/$1.40/1M

nvidia/nemotron-3-ultra-550b-a55b

128K ctx/$1.00/1M

nvidia/nemotron-3-nano-omni

300K ctx/$0.06/1M

deepseek-v4-pro-0424

164K ctx/$1.75/1M

moonshotai/kimi-k2.6

256K ctx/$0.95/1M

zai-org/glm-5.1

200K ctx/$1.40/1M

Reference

kimi-k3 questions

How much does kimi-k3 cost?

kimi-k3 is priced at $3.00 per million input tokens and $15.00 per million output tokens when accessed via Requesty. Those are the upstream provider rates: pay as you go adds 5% on top, or 0% if you bring your own provider keys.

What is the context window of kimi-k3?

kimi-k3 has a context window of 1.0M tokens, with a maximum output of 131K tokens per response. That's roughly 1,398 words of input you can fit in a single prompt.

What can kimi-k3 do?

kimi-k3 supports tool calling, extended reasoning. You can call it through any OpenAI-compatible client by pointing base_url to Requesty.

How do I use kimi-k3 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 "nebius/kimi-k3". The Quickstart above shows Python, JavaScript and cURL snippets.

Can I run kimi-k3 through Requesty?

Yes. kimi-k3 runs through Requesty's OpenAI-compatible API, served from Nebius AI. You do not host the model yourself: point base_url at Requesty, set the model to "nebius/kimi-k3", and requests are routed to the upstream provider with automatic failover. The same key gives you 600+ other models too.

Call kimi-k3 through one endpoint

One key for this endpoint and 600+ other models. No markup on provider prices, automatic failover, caching built in.

All Nebius AI models