Home
LLMs
Kimi K3

Kimi K3

Moonshot AI's flagship for long-horizon coding and knowledge work, successor to the Kimi K2 family: 2.8T parameters (104B active), Attention Residuals architecture, Kimi K3 License.

Params: 2.8T, 104B active
Context: 1M tokens
License: Kimi K3 License

Visit Moonshot AI Compare vs other LLMs →

Hosted API price

List price via Moonshot AI · per 1M tokens

USD

Input: $3.00 /1M
Output: $15.00 /1M
Cheapest: $2.00 / $12.00 via EcoHash
Hosted by: 8 providers

Hosted API pricing

How we price

Order

Providers are listed cheapest input rate first, then by output rate, a tie going to the creator's own listing. A provider that lists the model without a published rate sits at the end.

Prices

Base-tier, on-demand rates in USD per 1M tokens, converted at ECB reference rates where a provider publishes in another currency. Cached-input and batch rates show only where a provider publishes them. "Cheapest" marks the one provider with the lowest input and output pair; a tie wears no badge.

Cost

Input rate times the input tokens plus output rate times the output tokens, at the monthly volume set above the table. Cached-input, batch, long-context and reasoning-token billing are not modelled.

Transparency and funding

Affiliates: Affiliate links are marked. We may earn a commission if you click them, but commissions never affect the order.

Every provider serving Kimi K3, cheapest input first. Set your monthly volume to see what each would bill.

Input M Output M

Every provider serving Kimi K3, with its price per 1M input and output tokens and what the volume above would cost
Provider	Input /1M	Output /1M	Cached input /1M	Cost at 10M in + 2M out	Link
EcoHash Cheapest	$2.00	$12.00	$0.20	$44.00	Visit website
DigitalOcean	$2.85	$14.25	$0.29	$57.00	Visit website
Moonshot AI Creator	$3.00	$15.00	$0.30	$60.00	Visit website
EmpirioLabs AI	$3.00	$15.00		$60.00	Visit website
Lyceum	$3.00	$15.00	$0.75	$60.00	Visit website
Novita	$3.00	$15.00	$0.30	$60.00	Visit website
Together AI	$3.00	$15.00	$0.30	$60.00	Visit website
Azure	$3.30	$16.50	$0.33	$66.00	Visit website

Heads up: Base-tier, on-demand rates per 1M tokens; cached, batch and long-context tiers excluded. A provider may serve a shorter context or a quantized build than the creator's release. Verify before provisioning. More on how we price.

Estimated cost to self-host

Kimi K3 needs about 1,544 GB of GPU memory at 4-bit with 32K context. The cheapest rental that fits is 8× B300 at about $45,446 a month.

That costs the same as roughly 9B tokens a month on Moonshot AI's API. Self-hosting is more expensive below that volume.

Single node ≤ 8

What it costs to self-host Kimi K3 at each precision: the memory it needs, the cheapest rental that holds it, that rental's month and the API volume it pays off at
Precision	Memory	Cheapest fit (32K context)(Full 1M context)	Cost /mo	Break-even vs API
4-bit INT4 / FP4	1,544 GB1,598 GB	8× B300	$45,446$45,446	9B tokens /mo9B tokens /mo
8-bit FP8 / INT8	2,805 GB2,914 GB	– –	––	––
16-bit FP16 / BF16	5,605 GB5,714 GB	– –	––	––

Estimates based on median on-demand rates for Nvidia GPUs. Memory is weights plus KV cache for one request, using FP8 KV cache where supported and FP16/BF16 otherwise. Break-even assumes a 5:1 input-to-output ratio. No guarantee of runtime support, usable performance, or that a matching quantized build exists. How we estimate costs.

Similarly priced models

The models nearest Kimi K3 by blended rate, each at its own cheapest provider.

Models priced nearest Kimi K3 per 1M tokens, each at its cheapest provider
Model	Blended /1M	Input /1M	Output /1M	Context	Cutoff	vs Kimi K3
o3 OpenAI	$3.00	$2.00	$8.00	200K	Jun 2024	−18%
Claude Sonnet 5 Anthropic	$3.3333	$2.00	$10.00	1M	Jan 2026	−9%
GPT-5.6 Sol OpenAI	$3.3333	$2.00	$10.00	1M	Feb 2026	−9%
Gemini 3.1 Pro Preview Google Cloud	$3.6667	$2.00	$12.00	1M	Jan 2025	0%
GPT-5.6 Terra OpenAI	$3.6667	$2.00	$12.00	1M	Feb 2026	0%
Kimi K3 This model Moonshot AI	$3.6667	$2.00	$12.00	1M
Command A Cohere	$3.75	$2.50	$10.00	256K		+2%
Command R+ 08-2024 Cohere	$3.75	$2.50	$10.00	128K	Feb 2023	+2%
GPT-4o OpenAI	$3.75	$2.50	$10.00	128K	Oct 2023	+2%
GPT-5.2 OpenAI	$3.7917	$1.75	$14.00	400K	Aug 2025	+3%
GPT-5.3-Codex OpenAI	$3.7917	$1.75	$14.00	400K	Aug 2025	+3%

USD per 1M tokens at each model's cheapest listed provider. Blended is the cost of 10M input plus 2M output tokens, spread over the 12M.

Lyceum

NVIDIA H200 from $2.20/hr, B200 from $3.40/hr, B300 from $3.80/hr

Per-second billing with no minimum commitment or vendor lock-in

Scale to 8,000 GPUs over InfiniBand for distributed AI training

View Pricing

Capabilities

What Kimi K3 accepts and can do, as published by Moonshot AI.

Text Accepts and generates natural-language text.
Image Accepts images as input alongside text.
Video Accepts video as input.
Reasoning Thinks before it answers, always on or as a switchable mode.
Function calling Connect to external tools, APIs, and systems.
Structured output Return responses in structured formats like JSON.
Web search Search the web for up-to-date information.
Code execution Write and execute code in a sandboxed environment.
File upload Process uploaded documents, spreadsheets, and other files.

Common questions

What is Kimi K3 good for?

Long coding sessions and end-to-end knowledge work, which Moonshot calls its flagship for. It takes images and video, with weights under the custom Kimi K3 License.

When is Kimi K3 not a good fit?

Cost-sensitive or high-volume use: it sits at the top of Moonshot's price range. No audio input, and at 2.8T parameters self-hosting takes more than one node, under a custom licence.

What is the cheapest way to run Kimi K3?

Hosted, unless you push serious volume. Moonshot AI charges $3.00 in / $15.00 out per 1M tokens. The cheapest rental that fits is 8x B300 at $45,446 a month, which costs the same as about 9B tokens a month on that API.

Can I self-host Kimi K3?

Yes. The weights are released under the Kimi K3 License. At 4-bit it needs about 1,544 GB of GPU memory, which starts at roughly $45,446 a month on the cheapest rental that fits.