Provider profile: SiliconFlow - SiliconFlow's OpenAI-compatible API offers 200+ optimized models — DeepSeek R1, DeepSeek V3, Qwen3, and more. SiliconFlow integration is rolling out on TheRouter; some models (e.g. DeepSeek R1, Qwen3) are already routable directly today. TheRouter adds automatic failover, unified billing, and team governance on top.

← All providers

SiliconFlow

Global21 models

SiliconFlow's catalog of 200+ models — rollout in progress

SiliconFlow's OpenAI-compatible API offers 200+ optimized models — DeepSeek R1, DeepSeek V3, Qwen3, and more. SiliconFlow integration is rolling out on TheRouter; some models (e.g. DeepSeek R1, Qwen3) are already routable directly today. TheRouter adds automatic failover, unified billing, and team governance on top.

  • ✓OpenAI-compatible API — change one line (base_url) and your existing OpenAI SDK code works instantly
  • ✓3 models completely free — Qwen3-8B, DeepSeek-R1-Distill-Qwen-7B & DeepSeek-OCR at $0
  • ✓200+ optimized models including DeepSeek R1, Qwen, GLM, Kimi & more
  • ✓Privacy-first: no data stored, ever — ideal for sensitive workloads
OpenAI-compatible APIReasoningCodingVision

Quickstart

# Python (openai SDK)
from openai import OpenAI

client = OpenAI(
    base_url="https://api.therouter.ai/v1",  # only change needed
    api_key="YOUR_THEROUTER_KEY",
)

# Free model: Qwen3-8B via SiliconFlow — no credits needed
response = client.chat.completions.create(
    model="siliconflow/Qwen/Qwen3-8B",  # free model
    messages=[{"role": "user", "content": "Explain quantum entanglement"}],
    max_tokens=512,
)
print(response.choices[0].message.content)

# TypeScript / Node.js (openai SDK)
import OpenAI from "openai";
const client = new OpenAI({
  baseURL: "https://api.therouter.ai/v1",
  apiKey: "YOUR_THEROUTER_KEY",
});
const res = await client.chat.completions.create({
  model: "siliconflow/Qwen/Qwen3-8B",
  messages: [{ role: "user", content: "Explain quantum entanglement" }],
});
console.log(res.choices[0].message.content);

Use the global api.therouter.ai endpoint for new integrations; the legacy China accelerated endpoint is retired.

OpenAI-Compatible API

SiliconFlow via OpenAI SDK — zero code changes

Drop in DeepSeek, Qwen, and GLM with your existing OpenAI client. Change only baseURL. 3 models completely free.

Integration guide

Use with OpenAI SDK — Drop-in Compatible

SiliconFlow's API is fully OpenAI-compatible. Swap base_url and api_key — zero code changes. TheRouter handles failover, load balancing, and billing automatically.

Python

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_THEROUTER_API_KEY",
    base_url="https://api.therouter.ai/v1",
)

resp = client.chat.completions.create(
    model="deepseek/deepseek-v3-2",
    messages=[{"role": "user",
              "content": "Hello!"}],
)
print(resp.choices[0].message.content)

TypeScript

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.THE_ROUTER_API_KEY,
  baseURL: "https://api.therouter.ai/v1",
});

const resp = await client.chat
  .completions.create({
    model: "deepseek/deepseek-v3-2",
    messages: [{ role: "user",
      content: "Hello!" }],
  });
console.log(
  resp.choices[0].message.content
);

curl

curl https://api.therouter.ai/v1/chat/completions   -H "Content-Type: application/json"   -H "Authorization: Bearer $KEY"   -d '{
    "model": "deepseek/deepseek-v3-2",
    "messages": [
      {"role": "user",
       "content": "Hello!"}
    ]
  }'

Models

DeepSeek R1DeepSeek's flagship reasoning model with extended chain-of-thought
DeepSeek V3.2DeepSeek's latest chat model balancing speed and intelligence
DeepSeek V3.1Cost-effective chat model for high-throughput workloads
Qwen3 235BAlibaba's flagship MoE model
Qwen3 32BEfficient mid-size Qwen3 model
Qwen Coder 480BMassive coding-focused model
Qwen Coder 30BLightweight code generation specialist
Qwen3 8BFast and efficient small model

Frequently Asked Questions

Is SiliconFlow's API OpenAI-compatible? Can I use my existing OpenAI SDK?

Yes. SiliconFlow's API is fully OpenAI-compatible. Through TheRouter, you access SiliconFlow models by changing a single line: set base_url to https://api.therouter.ai/v1 and replace your OpenAI API key with a TheRouter key. No other code changes needed — streaming, function calling (tools), JSON mode, and stop sequences all work exactly as they do with OpenAI.

Which SiliconFlow models are free on TheRouter?

Three models are completely free: Qwen3-8B (general chat), DeepSeek-R1-Distill-Qwen-7B (reasoning), and DeepSeek-OCR (document extraction). No credit card required, no usage limits on the free tier.

Why use TheRouter for SiliconFlow instead of calling SiliconFlow directly?

TheRouter adds automatic failover (SiliconFlow fails → Bedrock takes over), a single API key across all providers, usage analytics, spend controls, and team governance — all without changing your code. If SiliconFlow has downtime, your requests are re-routed transparently.

What is SiliconFlow?

SiliconFlow is a high-performance AI inference platform with 200+ optimized models, OpenAI-compatible APIs, and a privacy-first approach — no data stored, ever. TheRouter integrates directly with their international platform at siliconflow.com and routes SiliconFlow models alongside all other providers under a unified API.

Help & contact