Model FatigueLLMs under $1, measured — and who holds your data
Models

kimi-k2.5

Kimi K2.5

Made by Moonshot AI. Sold by 3 providers, from $0.57 per million input tokens.

Maker
Moonshot AI
Released
2026-01-27
Context
262K tokens
Max output
262K tokens
Input
text, image, video
Output
text
Open weights
Yes
Knowledge cutoff
2026-01

Where to buy it

ProviderInput $/MOutput $/MBlended*Holds your promptTraining
Alibaba Cloud$0.57$3.01$1.18runs it 🇨🇳 Chinano training
Amazon Bedrock$0.60$3.00$1.20runs it 🇺🇸 United Statesno training
Microsoft Azure$0.60$3.00$1.20runs it 🇺🇸 United Statesno training

Blended is our calculation: (3 × input + output) ÷ 4, US dollars per million tokens — most calls read far more than they write. Tiers that are cheaper because they train on your data are not used for the “from” price.

Retirement dates

ProviderId on that providerShuts downSource
DeepInframoonshotai/Kimi-K2.52026-09-07provider page
Microsoft AzureKimi-K2.52027-01-26provider page

Compare with every other model →