RentGPU RentGPU
Trouver un GPU →
by Moonshot AI

MoonshotAI: Kimi K3.

multimodal closed 1M ctx
Cheapest input
$2.85/M
on DeepInfra
Cheapest output
$14.25/M
on DeepInfra
Hosted equiv.
~$5.13/hr
@ 100 tok/s on DeepInfra

Kimi K3 is a 2.8T parameter open-weight multimodal reasoning model from Moonshot AI. It is suited for complex coding, knowledge work, and long-horizon agentic workflows, and is particularly strong at...

Hosted API only

No self-host path — closed weights.

MoonshotAI: Kimi K3's weights aren't published. Use it via the access providers below.

Where to use it

Cheapest hosted endpoints.

Provider Access $/M in $/M out
DeepInfra hosted inference $2.85 $14.25 Launch ↗
Together AI hosted inference $3.0 $15.0 Launch ↗
OpenRouter api aggregator $3.0 $15.0 Launch ↗
FAQ

Frequently asked.

How do I run MoonshotAI: Kimi K3?
MoonshotAI: Kimi K3 is a closed-source API model. The cheapest way to access it is through the API providers listed on this page (direct API, aggregators, and hosted chat UIs).
Where can I access MoonshotAI: Kimi K3?
MoonshotAI: Kimi K3 is available via Fireworks AI, Together AI, DeepInfra, OpenRouter. Each access option lists its own pricing (per million tokens or hourly hosting).
How much does it cost to run MoonshotAI: Kimi K3?
API pricing starts at $3.0/M input tokens and $15.0/M output tokens. Self-hosting cost depends on the GPU you rent — see the Run It Yourself tab.
Is MoonshotAI: Kimi K3 open-source or proprietary?
MoonshotAI: Kimi K3 is a proprietary model from Moonshot AI. Access is API-only — there are no public weights to download.