MoonshotAIReleased on Sep 5, 2025

Kimi K2 0905: Benchmarks, Pricing & Context Window

Kimi K2 0905 is a language model from MoonshotAI, released in September 2025.

Kimi K2 0905 is the September update of Kimi K2 0711. It is a large-scale Mixture-of-Experts (MoE) language model developed by Moonshot AI, featuring 1 trillion total parameters with 32 billion active per forward pass. It supports

Input
Text
Output
Text
Speed
—
Cost
$0.81/ 1M · 8:1 in:out
$0.60 in · $2.50 out

Kimi K2 0905 pricing

Providers

Kimi K2 0905 starts at $0.600 per million input tokens and $2.50 per million output tokens via Novita.

ProviderInput $/MOutput $/MMax InputMax OutputLatency p95 sThroughput P95QuantInputOutput
Novita logoNovita
$0.600$2.50262.1K262.1K
—
—
fp8

Kimi K2 0905 API

API access coming soon

Kimi K2 0905 will be available through our gateway shortly.

Kimi K2 0905 examples

Recent arena outputs from Kimi K2 0905, picked from the highest-ranked matchups.

Kimi K2 0905 license

Kimi K2 0905 is released under the Proprietary license, which restricts commercial use, has 1.0T parameters.

License
Proprietary
Non-commercial
Parameters
1.0T

Proprietary license - usage restrictions apply

FAQ

Common questions about Kimi K2 0905.

When was Kimi K2 0905 released?

Kimi K2 0905 was released on September 5, 2025 by MoonshotAI. This is the official Kimi K2 0905 release date tracked on LLM Stats.

Is Kimi K2 0905 available via API?

Yes, Kimi K2 0905 is available via API. See the official documentation for authentication and endpoint details. It is served by 1 provider tracked on LLM Stats.

How big is Kimi K2 0905?

Kimi K2 0905 has 1000 billion parameters.

Who created Kimi K2 0905?

Kimi K2 0905 was created by MoonshotAI.

What is the license for Kimi K2 0905?

Kimi K2 0905 is released under the Proprietary license.

Where can I use Kimi K2 0905?

Kimi K2 0905 is available through 1 provider including Novita.

Where is the Kimi K2 0905 paper or technical report?

Kimi K2 0905 has a paper or technical report available at https://moonshot.cn/blog/kimi-k2-0905. Use that source for architecture, training, release and evaluation details.