Skip to main content
Watch: Stop Building Apps. Build Tools for AI Agents

Open-weight API pricing - last verified

Kimi K3: where to run it and what it costs

9 offers from 9 providers list Kimi K3 at a per-token price. The lowest listed input price is $2.85 per 1M input tokens at DeepInfra (Kimi-K3), with output at $14.25. First-party, Moonshot AI lists $3.00 input and $15.00 output per 1M tokens. Prices are as listed on October 18, 2026; they change often and are not a benchmark.

2.78T total, 104B active, mixture of experts. Native context 1,048,576 tokens, per our local-models record (from the Hugging Face model card).

Every offerQuotesRun it yourselfAll open-weight models

Offers
9
Providers
9
Lowest listed input
$2.85
First-party input
$3.00

Per-token prices

Every Kimi K3 offer, sorted by input price

Each row is one build as the provider lists it, with its own model ID: dated snapshots, Fast and turbo builds and regional rows are never merged. Sorted by listed input price in USD per 1M input tokens; every price is in that same unit. Not stated means the provider's page does not say.
9 hosted offers for Kimi K3, sorted by listed input price in USD per 1M input tokens
Provider and model IDInputUSD per 1M tokensOutputUSD per 1M tokensCached inputUSD per 1M tokensContext servedas the provider lists itQuantizationStatusSource
DeepInframoonshotai/Kimi-K3Quotes for this row$2.85$14.25$0.2851024kInferred from an unlabelled tagNot statedNot stateddeepinfra.com
Moonshot AI (first-party)kimi-k3First-partyQuotes for this row$3.00$15.00$0.301,048,576Not statedNot statedplatform.kimi.ai
Together AImoonshotai/Kimi-K3Listed as: Kimi K3Quotes for this row$3.00$15.00$0.301,048,576Not statedServerless Inferencetogether.ai
Fireworks AIkimi-k3Listed as: Kimi K3Quotes for this row$3.00$15.00$0.30Not statedNot statedStandarddocs.fireworks.ai
Basetenmoonshotai/Kimi-K3Listed as: Kimi K3Quotes for this row$3.00$15.00$0.301048kNot statedModel APIsbaseten.co
Novitamoonshotai/kimi-k3Listed as: Kimi K3Quotes for this row$3.00$15.00$0.301MNot statedNot statednovita.ai
Amazon BedrockNo API model ID on the pricing pageListed as: Kimi 3Inferred: matched by nameQuotes for this row$3.00$15.00$0.30Not statedNot statedGlobal CRISaws.amazon.com
Azure AI Foundry (Fireworks on Foundry)FW Kimi K3Quotes for this row$3.00$15.00$0.30Not statedNot statedGlobalazure.microsoft.com
Alibaba Cloud Model Studiokimi-k3Quotes for this row$3.00$15.00Not statedNot statedNot statedInternationalalibabacloud.com
First-party
The model's own developer selling its API.
Peak-hour price
DeepSeek's listed price during its peak hours; off-peak is half, per its page. Alibaba's busy-hour rows work the same way.
Card shows N% off
A discount tag sits beside the price on the provider's catalog card. The number is the discounted one and may be temporary.
Inferred
Our reading, not the page's words: a name-only model match, a region taken from table order, or a context figure from an unlabelled tag.
Below native
The provider serves a shorter context than the model's native window in our local-models data.
Preview
The provider lists the model as a preview, for evaluation rather than production.

Evidence

The quote behind every value

Each value is copied from the provider's own page; one quote that backs several values shows once.

DeepInfra

Kimi-K3

Model ID

“$0.285 cached, $2.85 in, $14.25 out / 1M”
deepinfra.com

Variant, Input, Output, Cached input

“Kimi-K3 1024k $2.85 / $0.285 cached $14.25”
deepinfra.com

Context served

“Kimi-K3 1024k”
deepinfra.com
  • Inferred from an unlabelled tag. Context inferred from an unlabelled tag: the quoted pricing row carries "1024k" with no Context label.

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Moonshot AI (first-party)

kimi-k3

Model ID, Variant, Input, Output, Cached input, Context served

“kimi-k3 1M tokens $3.00 $6.00 $0.30 $3.00 $15.00 1,048,576 tokens”
platform.kimi.ai

Input column, Output column, Cached input column

“Model Unit Cache Write Price (TTL 5min) Cache Write Price (TTL 1h) Cached Input Price Input Price Output Price Context Window”
platform.kimi.ai

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Source notes: Cache writes billed separately: $3.00 (5 min TTL) and $6.00 (1 h TTL) per 1M tokens.

Together AI

Kimi K3

Model ID

“Moonshot Kimi K3 moonshotai/Kimi-K3 1048576 $3.00 $0.30 $15.00 - Yes Yes”
docs.together.ai

Variant, Input, Output, Cached input

“Kimi K3 $3.00 $0.30 (cached) $15.00”
together.ai

Context served

“moonshotai/Kimi-K3 1048576”
docs.together.ai

Status

“Serverless Inference Most teams start with serverless inference”
together.ai

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization column shows "-" for this model.

Source notes: Together also sells Kimi K3 as Provisioned Throughput (PTU at $0.05 per PTU-minute); not a per-token price.

Fireworks AI

Kimi K3

Model ID

“app.fireworks.ai/models/fireworks/kimi-k3”
docs.fireworks.ai

Variant, Input, Output, Cached input

“Kimi K3 $3.00 / $0.30 / $15.00 $3.75 / $0.375 / $18.75”
docs.fireworks.ai

Input column, Output column, Cached input column

“prices are input / cached input / output (USD per 1M tokens), in that order”
docs.fireworks.ai

Status

“In each Standard or Priority cell, prices are input / cached input / output (USD per 1M tokens), in that order”
docs.fireworks.ai

Not stated

  • Batch: No per-model batch price listed; the page states batch inference is billed at 50% of serverless pricing.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Priority $3.75 / $0.375 / $18.75. Separate rows: Kimi K3 Fast and Kimi K3 (US), each $4.50 / $0.45 / $22.50 Standard; not merged.

Baseten

Kimi K3

Model ID, Context served

“Kimi K3 moonshotai/Kimi-K3 1048k 262k”
docs.baseten.co

Variant, Input, Output, Cached input

“Kimi K3 Kimi K3 $3.00 $3.00 $0.30 $0.30 $15.00”
baseten.co

Input column, Output column, Cached input column

“Price per 1M tokens Model Input Cache Input Output”
baseten.co

Status

“Model APIs Instant access to pre-optimized models running on the Baseten Inference Stack.”
baseten.co

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Quantization: Quantization not stated on the pages used.

Novita

Kimi K3

Model ID

“Kimi K3 moonshotai/kimi-k3”
novita.ai

Variant, Input, Output, Cached input

“Kimi K3 1M $3 /Mt· Cache Read$0.3 /Mt $15 /Mt”
novita.ai

Context served

“Kimi K3 1M $3”
novita.ai

Not stated

  • Batch: No per-model batch price listed; the pricing page states batch inference is at an introductory 50% discount for supported models.
  • Quantization: Model page lists Quantization as "-".
  • Status: No availability tier or release-stage label stated for this offer on the pages used.

Amazon Bedrock

Kimi 3

Variant, Status

“Moonshot AI On-Demand pricing - Kimi 3 - Kimi 2 - Kimi 3 Inference option Input Output Cache read Cache write (30 min) Global CRIS $3.00 $15.00 $0.30 $3.75”
aws.amazon.com

Input, Output, Cached input

“Kimi 3 Inference option Input Output Cache read Cache write (30 min) Global CRIS $3.00 $15.00 $0.30 $3.75”
aws.amazon.com
  • Inferred: matched by name. Amazon Bedrock lists this model as "Kimi 3" with no API model ID on the pricing page; it is matched to Kimi K3 by name only.

Not stated

  • Model ID: No API model ID string on the pricing page; the tab is labelled "Kimi 3".
  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Bedrock labels this model "Kimi 3"; matched to Kimi K3 by name only. US CRIS is $3.30 / $16.50 / $0.33 cache read; cache write (30 min) $3.75 Global. "Pricing shown is for the Standard tier."

Azure AI Foundry (Fireworks on Foundry)

FW Kimi K3

Model ID, Output

“FW Kimi K3 Global Input: $3/1M Tokens Cached Input: $0.30/1M Tokens Output: $15/1M Tokens”
azure.microsoft.com

Variant, Input, Status

“FW Kimi K3 Global Input: $3/1M Tokens”
azure.microsoft.com

Cached input

“FW Kimi K3 Global Input: $3/1M Tokens Cached Input: $0.30/1M Tokens”
azure.microsoft.com

Not stated

  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Data Zone deployment listed at $3.30 / $0.33 / $16.50.

Alibaba Cloud Model Studio

kimi-k3

Model ID, Variant, Input, Output, Status

“kimi-k3 International $3 $15”
alibabacloud.com

Input column, Output column

“Model ID Deployment scope Input price (per 1 million tokens) Output price (per 1 million tokens)”
alibabacloud.com

Not stated

  • Cached input: No cached-input price listed for this model on the pricing page.
  • Batch: No per-model batch price listed on the pricing page.
  • Context served: Served context length not stated on the pages used.
  • Quantization: Quantization not stated on the pages used.

Source notes: Global deployment scope rows list kimi-k3 at $2.827 / $14.133; not merged.

Coverage

Not listed on the pages read

  • Fireworks AI: only the headline-model price table is covered; models outside it fall under size-based pricing and are not listed on the pages read.
  • Cerebras: the pay-as-you-go Developer tier lists only GPT OSS 120B and Qwen 3.8 27B; other models are Dedicated Inference by quote.
  • Cloudflare Workers AI: no Kimi K3, DeepSeek V4.1 Flash, Gemma 4 31B, MiniMax-M3 or Nemotron 3.5 rows on the pages read.
  • Fireworks AI says batch inference is billed at 50% of serverless pricing, with no per-model batch figure.
  • Novita says batch inference is at an introductory 50% discount for supported models, with no per-model batch prices.
  • Together AI shows a Batch API price toggle but lists no per-model batch figures.
  • Amazon Bedrock says Flex tier and Batch pricing is at 50% discount to Standard tier pricing; per-model batch prices are listed only for gpt-oss, in the Asia Pacific (Sydney) table.

Alternatives

Run Kimi K3 yourself, or compare its first-party API

FAQ

How much does the Kimi K3 API cost?

Listed input prices run from $2.85 to $3.00 per 1M input tokens across 9 offers on October 18, 2026, and output from $14.25 to $15.00 per 1M output tokens. The lowest listed input price is at DeepInfra (Kimi-K3). First-party: Moonshot AI at $3.00 input.

Which providers serve Kimi K3?

Together AI, Fireworks AI, DeepInfra, Baseten, Novita AI, Amazon Bedrock, Azure AI Foundry, Alibaba Cloud Model Studio and Moonshot AI. Each lists one build, shown under the provider's own model name.

Do all providers serve Kimi K3's full context window?

Kimi K3's native context is 1,048,576 tokens in our local-models data. No provider that states a served context lists one clearly below that. Not stated by 4 offers.

Is there a batch price for Kimi K3?

No per-model batch price for Kimi K3 is listed on the pages read. Fireworks AI says batch inference is billed at 50% of serverless pricing, with no per-model batch figure. Novita says batch inference is at an introductory 50% discount for supported models, with no per-model batch prices. Together AI shows a Batch API price toggle but lists no per-model batch figures. Amazon Bedrock says Flex tier and Batch pricing is at 50% discount to Standard tier pricing; per-model batch prices are listed only for gpt-oss, in the Asia Pacific (Sydney) table.

Sources

Last verified from each provider's own pages (11 sites). This is a directory of listed prices and what each provider states about the model it serves, not a benchmark. Prices change often; check the provider's page before you commit. Speed figures appear only where a provider states them, labelled Vendor-stated.

Newsletter

Know when open-weight API prices move

One email per week. New model releases, price changes and where to run them, and what they mean for what you build.

Read more about Kimi K3

Related

Run open-weight models locally

The self-host alternative: downloads, memory needs and commands for each model.

Open-weight model licenses

Which of these models you may use commercially, and on what terms.

LLM API cost levers

Prompt caching, batch discounts and service tiers by provider.

Free LLM API tiers

Which providers have a standing free tier or credits.

OpenAI-compatible endpoints

Base URLs and limits for switching providers without rewriting code.

LLM gateways and routers

Route one model across several of these providers with fallbacks.

Other open-weight models: DeepSeek V4 Pro, DeepSeek V4.1 Flash, Qwen3.8-27B, GLM-5.3, MiniMax-M3, gpt-oss-120b, gpt-oss-20b, Llama 4 Maverick, Gemma 4 31B, Mistral Large 3, NVIDIA Nemotron 3.5 Super VL 120B-A12B.

Providers on this page: Together AI, Fireworks AI, DeepInfra, Baseten, Novita AI, Amazon Bedrock, Azure AI Foundry, Alibaba Cloud Model Studio, Moonshot AI.