Kimi K3 is developed by Moonshot AI. AIHubMix aggregates it alongside models from other providers behind one API and one bill.

Kimi K3

Compare

Kimi K3

Q: What is the context length of Kimi K3?

Kimi K3 has a 1,048,576 token context window. It supports up to 1,048,576 output tokens.

Q: How much does Kimi K3 cost?

On AIHubMix, Kimi K3 costs $2.84 per million input tokens and $14.2 per million output tokens. Cached input reads are billed at $0.284 per million tokens.

Q: What modalities does Kimi K3 support?

Kimi K3 accepts text, image and video input. It returns text output.

Q: What capabilities does Kimi K3 support?

Kimi K3 supports thinking, function calling and structured outputs. Per-protocol parameter support is listed in the capability table on this page.

Q: How do I call Kimi K3 via API?

Kimi K3 is available through the AIHubMix unified API. The API is OpenAI-compatible: point your OpenAI SDK at https://aihubmix.com/v1, use your AIHubMix API key, and set the model name to kimi-k3 — no other code changes needed.

Q: When was Kimi K3 released?

Kimi K3 was released on July 16, 2026 by Moonshot AI.

kimi-k3llms.txt

Moonshot AI

Kimi K3 is Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window and industry-leading intelligence.

Pricing

Input Tokens: $2.84 /M tokens
Output Tokens: $14.2 /M tokens
Cache Read: $0.284 /M tokens

Eingabemodalitäten

Text
Vision
Video

Ausgabemodalitäten

Text

Context length

1.05M tokens

Max output

1.05M tokens

Capabilities

Thinking
Streaming
Tool calling
Web search
URL context
Code interpreter
Computer use
File search
Memory tool
Structured outputs
Structured decision
Citations
Prompt caching
Background mode
Server-side sessions

Anbieter

Bytedance doubao-kimi-k3

Pricing$2.84$14.2

Cache Storage$0.0024/h/M tokens

Cache Read$0.284/M tokens

PrecisionⓘNot Disclosed

Latency21.9S

Throughput22.5TPS

Uptime

100.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

Baidu baidu-kimi-k3

Pricing$2.84$14.2

Cache$0.284

PrecisionⓘNot Disclosed

Latency4.3S

Throughput10.1TPS

Uptime

100.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

Deepinfra deepinfra-kimi-k3

Pricing$3.135$15.675

Cache$0.3135

PrecisionⓘMXFP4

Latency1.4S

Throughput20.6TPS

Uptime

100.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

AWS bedrock-kimi-k3

Pricing$3$15

Cache Read$0.3/M tokens

Cache Write$3.75/M tokens

PrecisionⓘNot Disclosed

Latency2.4S

Throughput8.0TPS

Uptime

0.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

Phala phala-kimi-k3

Pricing$3.135$15.675

Cache$0.3135

PrecisionⓘNot Disclosed

Latency0.8S

Throughput20.1TPS

Uptime

100.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

Moonshot moonshot-kimi-k3

Pricing$3$15

Cache$0.3

PrecisionⓘNot Disclosed

Latency3.1S

Throughput8.8TPS

Uptime

100.00% uptime 2 days ago

100.00% uptime yesterday

100.00% uptime today

Gmicloud gmicloud-kimi-k3

Pricing$3$15

Cache$0.3

PrecisionⓘNot Disclosed

Latency0.9S

Throughput73.1TPS

Uptime

0.00% uptime 2 days ago

0.00% uptime yesterday

0.00% uptime today

InferenceNet inferencenet-kimi-k3

Pricing$3.135$15.675

Cache$0.3135

PrecisionⓘFP4

Latency-

Throughput-

Uptime

0.00% uptime 2 days ago

0.00% uptime yesterday

0.00% uptime today

Performance for kimi-k3

Uptime ist der Prozentsatz der erfolgreichen Anfragen in den letzten 72 Stunden. AIHubMix überwacht jeden Anbieter kontinuierlich und wiederholt Anfragen automatisch beim nächstbesten Anbieter, wenn einer einen Fehler zurückgibt oder zu langsam antwortet; Latency ist die gesamte Umlaufzeit einer Anfrage (niedriger ist besser); Throughput gibt an, wie schnell das Modell schreibt (Tokens pro Sekunde, höher ist besser).

Uptime

Latency

Throughput

Try this model

Python

import os
from openai import OpenAI

client = OpenAI(
    api_key=os.environ["AIHUBMIX_API_KEY"],
    base_url="https://aihubmix.com/v1",
)

response = client.chat.completions.create(
    model="kimi-k3",
    messages=[
      {
        "role": "user",
        "content": "Hello, how are you?"
      }
    ],
    max_tokens=1024,
    stream=False,
)

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

Frequently asked questions

What is Kimi K3?

Kimi K3 is Kimi’s flagship model for long-horizon coding and end-to-end knowledge work, with a 1M-token context window and industry-leading intelligence.

Kimi K3

Pricing

Eingabemodalitäten

Ausgabemodalitäten

Context length

Max output

Capabilities

Anbieter

Performance for kimi-k3

Try this model

Frequently asked questions

What is Kimi K3?

What is the context length of Kimi K3?

How much does Kimi K3 cost?

What modalities does Kimi K3 support?

What capabilities does Kimi K3 support?

How do I call Kimi K3 via API?

Who created Kimi K3?

When was Kimi K3 released?

Compare Kimi K3

Produkt

Entwickler

Unternehmen

Bedingungen und Richtlinien

Kimi K3

Pricing

Eingabemodalitäten

Ausgabemodalitäten

Context length

Max output

Capabilities

Anbieter

Performance for kimi-k3

Try this model

Frequently asked questions

What is Kimi K3?

What is the context length of Kimi K3?

How much does Kimi K3 cost?

What modalities does Kimi K3 support?

What capabilities does Kimi K3 support?

How do I call Kimi K3 via API?

Who created Kimi K3?

When was Kimi K3 released?

Compare Kimi K3

More models from Moonshot AI

Produkt

Entwickler

Unternehmen

Bedingungen und Richtlinien