KuraberuAIupdated 16 Sept 2026

Kimi K2.5 pricing.

prices as of · 52 providers · 62 prices

Moonshot AI's Kimi K2.5 holds 262k tokens of context, reads images as well as text and has a reasoning mode. Open weights, so anyone may host it, which is why the prices below differ so much.

Kimi K2.5 costs $0.60 in and $3 out per million tokens from Moonshot AI; the cheapest price we can explain is $2.80 at NanoGPT, checked 16 Sept 2026.

tool callingreads imagesreasoningstructured outputopen weights262k contextreleased 27 Jan 2026
Compare withvs Qwen2.5 32B Instructvs QwQ Plusor any model →

Moonshot AI list price

$4.80/M tokens

$0.60 in · $3 out

read · 35% above the market median

Cheapest price we can explain

$2.80/M tokens

$0.30 in · $1.90 out · 42% under list

NanoGPT
View at NanoGPT ↗

read

What it costs for your work

pages a month

Moonshot AI list price

$0.54a month

cheapest we can explain · NanoGPT

$0.294a month

difference

$0.246saved

A page is about 600 tokens in and 60 out, a support reply 200 in and 300 out, an email 150 in and 250 out. Nothing else is counted: no cache discount, no batch rate, no free tier.

Every provider selling Kimi K2.5

Four groups, decided by the price and by what the provider publishes about it. Prices are per million tokens, and the share is that row against Moonshot AI's price.

The maker's list price

1 price

What the lab that trained the model charges on its own API.

  • Moonshot AI

    The price Moonshot AI charges for the model it trained. Every row below is measured against it.

    Moonshot AI ↗

    $0.60 / $3in / out per M

    1.00×of list price

Cheaper, and we know why

28 prices

Below list, with a published reason: hosting the open weights, a regional price list, credits, a stated discount.

  • NanoGPT

    Routes to a host serving the open weights; precision not disclosed.

    256k context hereNanoGPT ↗

    $0.30 / $1.90in / out per M

    0.58×of list price

  • NanoGPT

    Routes to a host serving the open weights; precision not disclosed.

    256k context hereNanoGPT ↗

    $0.30 / $1.90in / out per M

    0.58×of list price

  • LLM Gateway

    Routes to a host serving the open weights; precision not disclosed.

    LLM Gateway ↗

    $0.41 / $1.98in / out per M

    0.67×of list price

  • Avian

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    Avian ↗

    $0.45 / $2.20in / out per M

    0.74×of list price

  • OpenRouter

    Routes to a host serving the open weights; precision not disclosed.

    OpenRouter ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • SiliconFlow (China)

    SiliconFlow's China-region catalog, priced in the domestic market.

    262k context hereSiliconFlow (China) ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • DeepInfra

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    DeepInfra ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • SiliconFlow

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    262k context hereSiliconFlow ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • SiliconFlow

    Serves the open weights at int4, a lower precision than the maker's, so not quite the same product.

    int4 precisionSiliconFlow ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • Nous Portal

    Resells OpenRouter's catalog at 0.80× the listed price, funded by a subscription whose credits carry a 10% bonus.

    Nous Portal ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • Requesty

    Routes to a host serving the open weights; precision not disclosed.

    262k context hereRequesty ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • FastRouter

    Routes to a host serving the open weights; precision not disclosed.

    FastRouter ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • TrustedRouter

    Routes to a host serving the open weights; precision not disclosed.

    TrustedRouter ↗

    $0.47 / $2.32in / out per M

    0.78×of list price

  • Atlas Cloud

    Serves the open weights at int4, a lower precision than the maker's, so not quite the same product.

    int4 precisionAtlas Cloud ↗

    $0.49 / $2.50in / out per M

    0.83×of list price

  • Atlas Cloud

    Routes to a host serving the open weights; precision not disclosed.

    Atlas Cloud ↗

    $0.49 / $2.50in / out per M

    0.83×of list price

  • Alibaba Cloud (China)

    4% under the list price, an ordinary reseller margin.

    Alibaba Cloud (China) ↗

    $0.57 / $2.41in / out per M

    0.86×of list price

  • DigitalOcean Gradient

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    DigitalOcean Gradient ↗

    $0.50 / $2.70in / out per M

    0.88×of list price

  • Auriko

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    $0.50 / $2.80in / out per M

    0.90×of list price

  • TensorX

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    $0.50 / $2.80in / out per M

    0.90×of list price

  • Together AI

    Hosts the open weights on its own servers, so the price is its own, not a resale. Precision not disclosed.

    Together AI ↗

    $0.50 / $2.80in / out per M

    0.90×of list price

  • Eden AI

    Routes to a host serving the open weights; precision not disclosed.

    Eden AI ↗

    $0.50 / $2.80in / out per M

    0.90×of list price

  • Together AI

    Serves the open weights at fp4, a lower precision than the maker's, so not quite the same product.

    fp4 precisionTogether AI ↗

    $0.50 / $2.80in / out per M

    0.90×of list price

  • Cortecs

    Routes to a host serving the open weights; precision not disclosed.

    Cortecs ↗

    $0.51 / $2.87in / out per M

    0.92×of list price

  • Charm Hyper

    7% under the list price, an ordinary reseller margin.

    $0.56 / $2.94in / out per M

    0.96×of list price

  • io.net Intelligence

    7% under the list price, an ordinary reseller margin.

    io.net Intelligence ↗

    $0.56 / $2.94in / out per M

    0.96×of list price

showing the 25 cheapest of 28 · every row in the open dataset

Same price as the maker

25 prices

The same price the maker charges, sold by someone else.

  • AIHubMix

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • OrcaRouter

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Azure Cognitive Services

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Poe

    The same price Moonshot AI charges.

    128k context herePoe ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Novita

    The same price Moonshot AI charges.

    Novita ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Hugging Face Inference Providers

    The same price Moonshot AI charges.

    Hugging Face Inference Providers ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • DaoXE

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Jiekou

    The same price Moonshot AI charges.

    Jiekou ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • FrogBot

    The same price Moonshot AI charges.

    256k context here

    $0.60 / $3in / out per M

    1.00×of list price

  • Baseten

    The same price Moonshot AI charges.

    262k context hereBaseten ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Vercel AI Gateway

    The same price Moonshot AI charges.

    262k context hereVercel AI Gateway ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Jalapeno Cloud

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Kilo Gateway

    The same price Moonshot AI charges.

    Kilo Gateway ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Azure AI Foundry

    The same price Moonshot AI charges.

    Azure AI Foundry ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Merge Gateway

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Abacus

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • OpenCode Zen

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Ofox

    The same price Moonshot AI charges.

    Ofox ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • HPC-AI

    The same price Moonshot AI charges.

    256k context here

    $0.60 / $3in / out per M

    1.00×of list price

  • OpenCode Go

    The same price Moonshot AI charges.

    $0.60 / $3in / out per M

    1.00×of list price

  • Amazon Bedrock

    The same price Moonshot AI charges.

    Amazon Bedrock ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • Phala (RedPill)

    The same price Moonshot AI charges.

    Phala (RedPill) ↗

    $0.60 / $3in / out per M

    1.00×of list price

  • 302.AI

    10% above Moonshot AI's price.

    $0.66 / $3.30in / out per M

    1.10×of list price

  • Inference.net

    33% above Moonshot AI's price.

    Inference.net ↗

    $0.80 / $3.95in / out per M

    1.32×of list price

  • Zenifra

    400% above Moonshot AI's price.

    229k context hereZenifra ↗

    $3 / $15.70in / out per M

    5.15×of list price

Cheaper, and nobody says why

2 prices · never ranked

Below list on a closed model with nothing to account for it. Shown, never ranked, never a pick.

  • Meganova

    Lists closed models at 0.80× the maker's price across its catalog.

    Meganova ↗

    $0.45 / $2.25in / out per M

    0.75×of list price

  • Featherless

    We could not read this provider's prices reliably, so the row is listed but never ranked.

    Featherless ↗

    $0.80 / $3.40in / out per M

    1.21×of list price

Given away with limits

6 free tiers · never ranked

These providers give Kimi K2.5 away inside a quota of their own. A limit is not a price, so none of them counts as cheapest here or anywhere else on the site.

  • Alibaba Token Plan
  • Alibaba Coding Plan (China)
  • Alibaba Coding Plan
  • SCNet Token Plan
  • Tencent Coding Plan (China)
  • Alibaba Token Plan (China)

Frequently asked

Who sells Kimi K2.5 cheapest?
NanoGPT, at $0.30 in and $1.90 out per million tokens, read on 16 Sept 2026. Routes to a host serving the open weights; precision not disclosed.
Is Kimi K2.5 open weights?
Yes. Moonshot AI publishes the weights, so anyone may host Kimi K2.5, which is why 52 providers sell it and why their prices differ so much.
How much context does Kimi K2.5 hold?
262,144 tokens in one call, which is about 262k, and it can call your own functions.
How does Kimi K2.5 compare with the rest of the market?
At $4.80 per million tokens, Moonshot AI's price is 35% above the $3.55 median across the 337 text models we track.
More from Moonshot AIKimi K2.6Kimi K3Kimi K2.7 CodeKimi K2 0905Similar contextKimi K2.6Kimi K2.7 CodeGLM-5.1GLM-5