SiliconFlow API models & prices

SiliconFlow Verified 30 days ago · 2026-10-14 API platform

1 plan

SiliconFlow is a multi-model API provider and inference platform. It serves models developed by DeepSeek, Alibaba (Qwen), Z.ai (GLM), Moonshot AI (Kimi), MiniMax and OpenAI. Text API calls are billed in USD per million tokens using each model’s published rates. These are SiliconFlow international service quotes, separate from model-developer prices and from the China market. No public monthly subscription is verified. A new-account $1 trial balance is advertised, with detailed claiming conditions and expiry unpublished; check the official account console.

Found a data issue? Report it·Reports go through manual verification and never change published prices directly.·Claim this product

All plans · 1

API models and prices · 11

Usage-based billing: text rates are usually per million tokens; other units and tiers are stated per entry. Compare only the same currency, unit and billing conditions. Unverified cache pricing does not mean free.

DeepSeek-V4.1-Flash

Official model ID: deepseek-ai/DeepSeek-V4.1-Flash

Model developer: DeepSeek · Service provider: SiliconFlow · Market: com

Model family: DeepSeek-V4.1-Flash · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.15
OutputUSD 0.6
Cached readUSD 0.003

Official SiliconFlow international API rates in USD per million tokens. Model developer: DeepSeek; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 393,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

DeepSeek-V4-Pro

Official model ID: deepseek-ai/DeepSeek-V4-Pro

Model developer: DeepSeek · Service provider: SiliconFlow · Market: com

Model family: DeepSeek-V4-Pro · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 1.50162
OutputUSD 3.135
Cached readUSD 0.135

Official SiliconFlow international API rates in USD per million tokens. Model developer: DeepSeek; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 393,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

Qwen3.8-2.4T-A95B

Official model ID: Qwen/Qwen3.8-2.4T-A95B

Model developer: Alibaba · Service provider: SiliconFlow · Market: com

Model family: Qwen3.8-2.4T-A95B · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 2
OutputUSD 6
Cached readUSD 0.25

Official SiliconFlow international API rates in USD per million tokens. Model developer: Alibaba; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 131,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

Qwen3.6-27B

Official model ID: Qwen/Qwen3.6-27B

Model developer: Alibaba · Service provider: SiliconFlow · Market: com

Model family: Qwen3.6-27B · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.3
OutputUSD 3.2
Cached readUnverified

The official pricing page does not publish cached input rates. Unknown rates are not zero.

Documented API context: 262,000 tokens · Official rounded display; exact token count unverified

Maximum output: 262,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

GLM-5.3

Official model ID: zai-org/GLM-5.3

Model developer: Z.ai · Service provider: SiliconFlow · Market: com

Model family: GLM-5.3 · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 1.4
OutputUSD 4.4
Cached readUSD 0.26

Official SiliconFlow international API rates in USD per million tokens. Model developer: Z.ai; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 262,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

GLM-5.3-Flash

Official model ID: zai-org/GLM-5.3-Flash

Model developer: Z.ai · Service provider: SiliconFlow · Market: com

Model family: GLM-5.3-Flash · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.15
OutputUSD 0.5
Cached readUSD 0.03

Official SiliconFlow international API rates in USD per million tokens. Model developer: Z.ai; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 262,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

GLM-5.1

Official model ID: zai-org/GLM-5.1

Model developer: Z.ai · Service provider: SiliconFlow · Market: com

Model family: GLM-5.1 · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 1.19
OutputUSD 3.74
Cached readUSD 0.6

Official SiliconFlow international API rates in USD per million tokens. Model developer: Z.ai; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 205,000 tokens · Official rounded display; exact token count unverified

Maximum output: 131,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

Kimi-K3

Official model ID: moonshotai/Kimi-K3

Model developer: Moonshot AI · Service provider: SiliconFlow · Market: com

Model family: Kimi-K3 · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 2.7
OutputUSD 13.5
Cached readUSD 0.27

Official SiliconFlow international API rates in USD per million tokens. Model developer: Moonshot AI; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 262,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

Kimi-K2.7-Code

Official model ID: moonshotai/Kimi-K2.7-Code

Model developer: Moonshot AI · Service provider: SiliconFlow · Market: com

Model family: Kimi-K2.7-Code · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.85916
OutputUSD 3.8
Cached readUSD 0.17993

Official SiliconFlow international API rates in USD per million tokens. Model developer: Moonshot AI; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 262,000 tokens · Official rounded display; exact token count unverified

Maximum output: 262,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

MiniMax-M3

Official model ID: MiniMaxAI/MiniMax-M3

Model developer: MiniMax · Service provider: SiliconFlow · Market: com

Model family: MiniMax-M3 · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.3
OutputUSD 1.2
Cached readUSD 0.06

Official SiliconFlow international API rates in USD per million tokens. Model developer: MiniMax; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 1,049,000 tokens · Official rounded display; exact token count unverified

Maximum output: 131,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

gpt-oss-120b

Official model ID: openai/gpt-oss-120b

Model developer: OpenAI · Service provider: SiliconFlow · Market: com

Model family: gpt-oss-120b · Currency: USD

Rates per million tokens
Price dimensionCurrent rate
InputUSD 0.05
OutputUSD 0.45
Cached readUSD 0.075

Official SiliconFlow international API rates in USD per million tokens. Model developer: OpenAI; inference service provider: SiliconFlow. Unknown cache charges are not zero.

Documented API context: 131,000 tokens · Official rounded display; exact token count unverified

Maximum output: 8,000 tokens · Official rounded display

Requires a SiliconFlow international account and API key. Per-account RPM/TPM and payment eligibility must be checked in the console.

Official source · Last verified: 2026-10-14T00:46:37.760Z

Need a recommendation?

The AI advisor considers your budget, tools, usage and purchase conditions to suggest primary and backup options.

Open AI advisor →