Moonshotai: Kimi K3

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 navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.

Model specifications

Input
text, image
Output
text
Context
1,048,576 tokens
Max output
1,048,576 tokens
Input price
.5 / 1M tokens
Output price
$7.5 / 1M tokens
Released
2026-07-17

Capabilities

  • Streaming
  • Function calling
  • Vision
  • JSON mode

Provider pricing, discounts and data privacy

Compare effective provider prices, published discounts, regions, retention policies, training use, compliance, and privacy links by service tier.

Flex service tier

1 available provider · discounts up to 50% · tier input average .4135 / 1M tokens · tier output average $7.0665 / 1M tokens

Alibaba Cloud Int.(CN)

Tier: Flex · Region: CN · Quantization: int4

Pricing
Input
.4135 / 1M tokens
Output
$7.0665 / 1M tokens
Cache read
$0.1415 / 1M tokens
Discounts
  • Input: 50% off — list $2.827 / 1M tokens, discounted .4135 / 1M tokens, effective .4135 / 1M tokens
  • Cache read: 50% off — list $0.283 / 1M tokens, discounted $0.1415 / 1M tokens, effective $0.1415 / 1M tokens
  • Output: 50% off — list 4.133 / 1M tokens, discounted $7.0665 / 1M tokens, effective $7.0665 / 1M tokens
Data privacy and compliance
Region
CN
Zero data retention
Yes
Data retention
60-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
Yes

Standard service tier

7 available providers · discounts up to 50% · tier input average $2.105786 / 1M tokens · tier output average 0.5285 / 1M tokens

Alibaba Cloud Int.(EU)

Tier: Standard · Region: EU · Quantization: int4

Pricing
Input
.4135 / 1M tokens
Output
$7.0665 / 1M tokens
Cache read
$0.1415 / 1M tokens
Discounts
  • Input: 50% off — list $2.827 / 1M tokens, discounted .4135 / 1M tokens, effective .4135 / 1M tokens
  • Cache read: 50% off — list $0.283 / 1M tokens, discounted $0.1415 / 1M tokens, effective $0.1415 / 1M tokens
  • Output: 50% off — list 4.133 / 1M tokens, discounted $7.0665 / 1M tokens, effective $7.0665 / 1M tokens
Data privacy and compliance
Region
EU
Zero data retention
Yes
Data retention
30-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
No

Alibaba Cloud Int.(HK)

Tier: Standard · Region: HK · Quantization: int4

Pricing
Input
.4135 / 1M tokens
Output
$7.0665 / 1M tokens
Cache read
$0.1415 / 1M tokens
Discounts
  • Input: 50% off — list $2.827 / 1M tokens, discounted .4135 / 1M tokens, effective .4135 / 1M tokens
  • Cache read: 50% off — list $0.283 / 1M tokens, discounted $0.1415 / 1M tokens, effective $0.1415 / 1M tokens
  • Output: 50% off — list 4.133 / 1M tokens, discounted $7.0665 / 1M tokens, effective $7.0665 / 1M tokens
Data privacy and compliance
Region
HK
Zero data retention
Yes
Data retention
30-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
No

Alibaba Cloud Int.(JP)

Tier: Standard · Region: JP · Quantization: int4

Pricing
Input
.4135 / 1M tokens
Output
$7.0665 / 1M tokens
Cache read
$0.1415 / 1M tokens
Discounts
  • Input: 50% off — list $2.827 / 1M tokens, discounted .4135 / 1M tokens, effective .4135 / 1M tokens
  • Cache read: 50% off — list $0.283 / 1M tokens, discounted $0.1415 / 1M tokens, effective $0.1415 / 1M tokens
  • Output: 50% off — list 4.133 / 1M tokens, discounted $7.0665 / 1M tokens, effective $7.0665 / 1M tokens
Data privacy and compliance
Region
JP
Zero data retention
Yes
Data retention
30-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
No

Alibaba Cloud Int.(SG)

Tier: Standard · Region: SG · Quantization: int4

Pricing
Input
.5 / 1M tokens
Output
$7.5 / 1M tokens
Cache read
$0.15 / 1M tokens
Discounts
  • Input: 50% off — list $3 / 1M tokens, discounted .5 / 1M tokens, effective .5 / 1M tokens
  • Cache read: 50% off — list $0.3 / 1M tokens, discounted $0.15 / 1M tokens, effective $0.15 / 1M tokens
  • Output: 50% off — list 5 / 1M tokens, discounted $7.5 / 1M tokens, effective $7.5 / 1M tokens
Data privacy and compliance
Region
SG
Zero data retention
Yes
Data retention
30-day retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
Yes
HIPAA compliant
No
SOC 2 certified
Yes
BYOK supported
Yes

Fireworks

Tier: Standard · Region: US

Pricing
Input
$3 / 1M tokens
Output
5 / 1M tokens
Cache read
$0.3 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
US
Zero data retention
Yes
Data retention
Zero retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
No
HIPAA compliant
No
SOC 2 certified
No
BYOK supported
No

Moonshot AI

Tier: Standard · Region: SG · Quantization: int4

Pricing
Input
$3 / 1M tokens
Output
5 / 1M tokens
Cache read
$0.3 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
SG
Zero data retention
No
Data retention
Unknown retention
Used for training
Yes
Data collection
Moderated
No
GDPR compliant
No
HIPAA compliant
No
SOC 2 certified
No
BYOK supported
No

Together

Tier: Standard · Region: US

Pricing
Input
$3 / 1M tokens
Output
5 / 1M tokens
Cache read
$0.3 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
US
Zero data retention
No
Data retention
Unknown retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
No
HIPAA compliant
No
SOC 2 certified
No
BYOK supported
No

Priority service tier

1 available provider · tier input average $3.75 / 1M tokens · tier output average 8.75 / 1M tokens

Fireworks

Tier: Priority · Region: US

Pricing
Input
$3.75 / 1M tokens
Output
8.75 / 1M tokens
Cache read
$0.375 / 1M tokens

No provider discount is currently published.

Data privacy and compliance
Region
US
Zero data retention
No
Data retention
Zero retention
Used for training
No
Data collection
Moderated
No
GDPR compliant
No
HIPAA compliant
No
SOC 2 certified
No
BYOK supported
No

Frequently asked questions

What is 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 navigating large repositories, using tools, debugging, and iterating against images, logs, tests, and runtime feedback. Its architecture uses KDA and Attention Residuals for computational efficiency.
How much does Moonshotai: Kimi K3 cost?
Input costs start at .5 / 1M tokens and output costs start at $7.5 / 1M tokens. Provider-level prices vary by service tier.
Are provider discounts available for Moonshotai: Kimi K3?
Yes. Current provider offers include discounts of up to 50% from list price. The provider table shows list, discounted, and effective prices.
What is the context length of Moonshotai: Kimi K3?
Moonshotai: Kimi K3 supports a 1,048,576 token context window and up to 1,048,576 output tokens.
What capabilities does Moonshotai: Kimi K3 support?
Moonshotai: Kimi K3 supports Streaming, Function calling, Vision, JSON mode.
Which providers offer Moonshotai: Kimi K3?
Moonshotai: Kimi K3 is available from Alibaba Cloud Int.(CN), Alibaba Cloud Int.(EU), Alibaba Cloud Int.(HK), Alibaba Cloud Int.(JP), Alibaba Cloud Int.(SG), Fireworks, Moonshot AI, Together.
How do providers handle data privacy for Moonshotai: Kimi K3?
8 of 8 providers report that customer data is not used for training, and 5 offer zero-data-retention routing. Retention, compliance, and privacy-policy links are listed per provider.

Browse all AI models