Kimi K3 is Moonshot AI's 2.8-trillion-parameter native multimodal model for software engineering, knowledge work, and deep reasoning. It uses Kimi Delta Attention, Attention Residuals, and a Stable LatentMoE architecture that activates 16 of 896 experts, accepts text, image, and video input, and supports a 1,048,576-token context window.

What is the context length of Kimi K3?

Kimi K3 has a recorded input context length of 1,048,576 tokens.

What providers serve Kimi K3, and can I use it via API?

Kimi K3 is available through the Phaseo API, with 16 active providers currently recorded. Recorded providers include Abacus, Arcee AI, AtlasCloud, Baseten, Berget.AI, Canopy Wave, CrofAI, and CrossModel and 30 more.

Does Kimi K3 support tool calling?

Yes. At least one active provider route for Kimi K3 currently advertises tool calling support. Provider support can vary, so requests are routed only to compatible routes.

Does Kimi K3 support structured outputs?

Yes. At least one active provider route for Kimi K3 currently advertises structured outputs support. Provider support can vary, so requests are routed only to compatible routes.

Kimi K3 API Pricing — Compare 58 Providers | Phaseo

Name: Moonshot Kimi K3
Creator: Moonshot

Providers

API providers, route pricing, availability, and recent reliability signals.

Performance

Latency, throughput, and reliability signals from recent traffic.

Pricing

Effective prices over the last 30 days, with current provider list prices for context.

Benchmarks

Activity

Daily gateway activity over the last 30 days, with current UTC-day pace projection.

Apps Using This Model

Public apps observed in gateway usage for this model.

Model Uptime

Uptime trend for this model over the last 24 hours.

About

Key dates, capabilities, and model metadata.

Subscriptions

Commercial plans and bundled access that currently include this model.

Kimi K3 API Pricing — Compare 58 Providers | Phaseo | Phaseo

Providers

API providers, route pricing, availability, and recent reliability signals.


CrofAI	$2	$8	$0.25	--	--	--

Morph	$2.8	$14	$0.29	--	--	--

DeepInfra	$2.85	$14.25	$0.285	--	--	--

Moonshot AI Standard/Fast	$3	$15	$0.3	--	--	--

Arcee AI	$3 image $0	$15	$0.3	--	--	--

AtlasCloud	$3	$15	$0.3	--	--	--

Baseten	$3 image $0	$15	$0.3	--	--	--

Canopy Wave	$3	$15	$0.3	--	--	--

Fireworks	$3	$15	$0.3	--	--	--

GMICloud	$3	$15	$0.3	--	--	--

Nebius Token Factory	$3	$15	--	--	--	--

NovitaAI	$3	$15	$0.3	--	--	--

SiliconFlow	$3	$15	$0.3	--	--	--

TensorX	$3	$15	$0.75	--	--	--

Together	$3	$15	$0.3	--	--	--

Wafer	$3	$15	$0.3	--	--	--

IO.net	$3.86	$17.8	$1.93	--	--	--

Venice	$3.75	$18.75	$0.375	--	--	--

Weights & Biases	$3	$15	$0.3	--	--	--

DigitalOcean	$2.85	$14.25	$0.285	--	--	--

NVIDIA	--	--	--	--	--	--

Performance

Hourly latency and throughput from the past seven days.

No gateway telemetry in the past 7 days

Hourly charts will appear when this model processes gateway traffic in the selected location.

Pricing

Weighted provider pricing over the last 30 days, with recent route pricing history below.

Effective pricing

Weighted by observed usage over the last 30 days. Every provider with recorded pricing is included.

Weighted input price

Per 1M tokens

Weighted output price

Per 1M tokens

Price history

USD per 1M tokens · 30-day rolling effective prices · exact UTC list changes


Moonshot AI	--	--	$3	$15	34.8%
Arcee AI	--	--	$3	$15	--
AtlasCloud	--	--	$3	$15	--
Baseten	--	--	$3	$15	--
Canopy Wave	--	--	$3	$15	--
CrofAI	--	--	$2	$8	--
DeepInfra	--	--	$2.85	$14.25	--
DigitalOcean	--	--	$2.85	$14.25	--
Fireworks	--	--	$3	$15	--
GMICloud	--	--	$3	$15	--
IO.net	--	--	$3.86	$17.8	--
Moonshot AI Turbo priority	--	--	--	--	--
Morph	--	--	$2.8	$14	--
Nebius Token Factory	--	--	$3	$15	--
NovitaAI	--	--	$3	$15	--
NVIDIA	--	--	--	--	--
SiliconFlow	--	--	$3	$15	--
TensorX	--	--	$3	$15	--
Together	--	--	$3	$15	--
Venice	--	--	$3.75	$18.75	--
Wafer	--	--	$3	$15	--
Weights & Biases	--	--	$3	$15	--