SiliconFlow gateway details

A China-based model API platform offering DeepSeek, Qwen, image, speech, and vector capabilities; ideal for developers who prioritize domestic models and low-latency nodes.

Best for: For Chinese apps needing domestic LLMs, vector retrieval, speech, and image APIs.

Current status

Reachable

Latency reference

380 ms

/models check

Model hub supported

Last verified

2026-07-01

Status notes

A Chinese model platform, good for DeepSeek, Qwen, embeddings, speech, and image capabilities.

Pricing mode

Billed per model and token usage; some models have free or low-cost tiers.

Price tags

Low cost, Signup credit, Pay-as-you-go

Free quota

New users or some models may have free quota; refer to the official console.

Usually you create an API key in the platform console, then replace the SDK's baseURL, model name, and auth details per the docs. Test streaming, function calling, error codes, and rate limits before launch.

import OpenAI from "openai";

const client = new OpenAI({
  apiKey: process.env.AI_GATEWAY_API_KEY,
  baseURL: "see the platform docs"
});

const response = await client.chat.completions.create({
  model: "DeepSeek",
  messages: [{ role: "user", content: "Test one call through an AI gateway." }]
});

• Supported models: DeepSeek, Qwen, Moonshot, Flux, Embedding, TTS, Whisper.
• Capabilities: OpenAI-compatible, Streaming, Function calling, Image generation, Speech-to-text, TTS, Embeddings, Balance query.
• Documentation completeness: High; editorial stability score 4.2 / 5.

• Not a relay for overseas closed-source models; if you need Claude or Gemini, pair it with another gateway.
• Third-party relay pricing, models, and availability can change, so re-check them periodically.

Is SiliconFlow compatible with the OpenAI API?

This entry is marked as supporting the OpenAI-compatible format, but the exact base_url, model names, and parameters still follow the official docs.

Is SiliconFlow ready for production?

Run a small-traffic load test, rate-limit test, and error-code test first, and confirm the billing rules; keep a backup provider for important workloads.