Kimi K3

by MoonshotAI · moonshotai/kimi-k3

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...

Input / 1M tokens
$0.192
Output / 1M tokens
$15.00
Context
1.05M

Specs

Input price$0.192 / 1M tokens
Output price$15.00 / 1M tokens
Cached input$0.182 / 1M tokens
Context window1.05M tokens
Max output944K tokens
Readstext, image, video
Writestext
Tool callingYes
ReasoningYes
ListedJul 16, 2026

List prices in USD, updated Oct 30, 2026. Providers may charge differently.

Providers

30 providers serve this model, as of Oct 30, 2026, 18:12 UTC.

Providers serving this model, sorted by price
ProviderQuantizationContextMax outputInput / 1MOutput / 1MCache read / 1MUptime (30 min)Latency (median)Throughput (median)SupportsStatus
InferenceNetfp41.05M944K$0.234$6.97$0.195100%788 ms72 tok/sNormal
InferenceNetfp41.05M944K$0.234$6.97$0.195100%882 ms46 tok/sNormal
Relacefp41.05M944K$0.153$9.75$0.15100%6.04 s21 tok/sNormal
Wafermxfp41.05M944K$0.276$9.94$0.13100%670 ms64 tok/sNormal
Sail Researchfp41.05M944K$0.23$11.00$0.115100%1.08 s55 tok/sNormal
Sail Researchfp41.05M944K$0.23$11.00$0.115100%1.55 s58 tok/sNormal
Chutesmxfp41.05M65.5K$1.85$9.50$0.185—3.47 s17 tok/sNormal
Decartmxfp41.05M944K$1.95$9.75$0.195100%5.89 s37 tok/sNormal
InferenceNetfp41.05M944K$0.50$11.88$0.35100%978 ms58 tok/sNormal
InferenceNetfp41.05M944K$0.50$11.88$0.35100%900 ms59 tok/sNormal
Fireworks—1.05M944K$2.10$10.50$0.21100%1.15 s55 tok/sNormal
Makorafp41.05M944K$0.50$13.00$0.20100%1.18 s53 tok/sNormal
Wafermxfp41.05M944K$0.192$15.00$0.182100%803 ms60 tok/sNormal
AkashMLFastestfp41.05M944K$1.20$14.00$1.20—658 ms141 tok/sNormal
Phala—1.05M944K$2.55$12.75$0.255—6.13 s42 tok/sNormal
DigitalOcean—1.05M944K$2.55$12.95$0.255100%1.81 s19 tok/sNormal
Morphfp81.05M944K$1.28$14.63$0.36100%2.21 s37 tok/sNormal
DeepInframxfp41.05M944K$2.85$14.25$0.28596.7%884 ms61 tok/sNormal
Amazon Bedrock—1.05M128K$3.00$15.00$0.30100%3.13 s37 tok/sNormal
BaseTenfp81.05M262K$3.00$15.00$0.30100%1.21 s48 tok/sNormal
Fireworks—1.05M944K$3.00$15.00$0.30100%3.61 s62 tok/sNormal
Modalmxfp41.05M944K$3.00$15.00$0.30100%2.05 s40 tok/sNormal
Moonshot AI—1.05M944K$3.00$15.00$0.30100%7.15 s26 tok/sNormal
Parasailfp41.05M944K$3.00$15.00$0.30100%1.13 s88 tok/sNormal
ParetoAlpha—1.05M944K$3.00$15.00$0.30—11.78 s15 tok/sNormal
Together—1.05M944K$3.00$15.00$0.30100%1.02 s50 tok/sNormal
Alibaba—1.05M131K$3.45$17.25$0.345100%2.15 s32 tok/sNormal
Wafermxfp41.05M944K$0.161$22.00$0.16100%701 ms63 tok/sNormal
Fireworks—1.05M944K$4.50$22.50$0.45—712 ms44 tok/sNormal
Fireworks—1.05M944K$4.50$22.50$0.45100%1.13 s68 tok/sNormal

"Cheapest" is the lowest input plus output price; "Fastest" the highest median throughput. Columns empty for every provider are not shown.

Usage

#23
by tokens, Oct 29, 2026
273B
tokens that day · 7 days −4%

Data as of 2026-10-30T18:12:03.879Z.All rankings