SF

SiliconFlow

online

China's fastest open-source inference. 14M tokens free.

LLMest. 2023 · Beijing, China
87
/ 100 APIVault Score

// At a glance

Free Tier
14M tokens free · no card
Category
LLM
Credit Card
Not required
Last Verified
2m ago

// Free tier details

Available Models

Qwen2.5 72BDeepSeek-V3GLM-4Llama 3.3 70B

Monthly Tokens

14M tokens on signup

No credit card needed
Phone verification required

// Quick start

300">"text-purple-400">from openai 300">"text-purple-400">import OpenAI

client = OpenAI(
    api_key=300">"YOUR_SILICONFLOW_KEY",
    base_url=300">"https://api.siliconflow.cn/v1",
)

response = client.chat.completions.create(
    model=300">"Qwen/Qwen2.5-72B-Instruct",
    messages=[{300">"role": 300">"user", 300">"content": 300">"Hello."}],
)

print(response.choices[0].message.content)

// Overview

SiliconFlow (硅基流动) is the leading Chinese inference platform serving Qwen, DeepSeek, and international models. New accounts receive 14M free tokens — one of the most generous free tiers in China.

// Pros

  • 14M free tokens is very generous
  • Access to top Chinese + international models
  • OpenAI-compatible

// Cons

  • Phone verification required
  • Chinese-language docs primarily
  • May require VPN outside China

// Score breakdown

Reliability (35%) (from 2m ago health check)100/100
Free Tier Generosity (30%) (computed from quota, no-CC, no-phone fields)75/100
Documentation (20%) (human rating)82/100
Popularity (15%) (GitHub stars (log-normalised), or manual baseline)84/100

Methodology: apivault.dev/methodology

// Best for

Chinese language AIQwen model accessCost-effective inference in Asia

// Recent changes

May 1, 2026Increased free signup tokens to 14Mfree tier changed