← LLM API prices

Kimi K3

Moonshot

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...

In
$2.65
Out
$13.28
Blended
$5.31
Where to buy this model
17 access routes
RouteInOutBlendedvs
Morph cheapestfp8$1.88$10.50$4.03-
InferenceNet $2.10$10.95$4.31+7%
Relace fp4$2.40$12.00$4.80+19%
Makora $2.55$12.75$5.10+27%
DigitalOcean $2.55$12.95$5.15+28%
Sail Research fp4$2.65$13.28$5.31+32%
Phala $2.70$13.50$5.40+34%
Wafer $3.00$12.75$5.44+35%
DeepInfra bf16$2.85$14.25$5.70+41%
Chutes mxfp4$3.00$15.00$6.00+49%
Parasail fp4$3.00$15.00$6.00+49%
Modal mxfp4$3.00$15.00$6.00+49%
Together $3.00$15.00$6.00+49%
Fireworks $3.00$15.00$6.00+49%
BaseTen fp8$3.00$15.00$6.00+49%
Moonshot AI mxfp4$3.00$15.00$6.00+49%
Alibaba $3.45$17.25$6.90+71%

official API · cloud platform · inference host · USD per 1M tokens. Same model, different providers - price and quantization vary.

Ctx
1.048576M
max output
944K
cache read /1M
$0.30
Family
kimi
Modalities
text · image · video
Features
json · reasoning · tools

Live pricing, updated 2026-09-15.

Concepts on this page