Whether you're a developer or a beginner, you'll find the perfect way to integrate here.
Supports OpenAI SDK, REST API, and browser direct calls β choose any of the three.
FlowerWolf is an AI Model Aggregation Platform β we integrate the best AI models globally (OpenAI, Google, Anthropic, domestic models, and more) into one place. With just one API Key, you can call all models.
Official channels are expensive, top-ups are cumbersome, and customer support is slow. We offer simpler integration, more stable service, and transparent pricing. Sign up and get trial credits β try before you decide.
Choose the method that suits you best
Use the ready-made OpenAI SDK β integrate in just 2 lines of code, and access top global models directly.
EasiestDirect HTTP requests β no language restrictions, works with Python/Java/JS/Shell.
Most UniversalHave GPUs and servers? We provide OpenClaw and Hermes deployment tools to run models yourself.
Full ControlNo GPU? We have H100/A100 compute available for rent, pay-per-use, flexible.
Pay-as-you-goSupports OpenAI official SDK β just change two lines of config
import openai
openai.api_key = "your-flowerwolf-api-key"
openai.base_url = "https://api.flowerwolf.net/v1"
# That's it β start using it!
response = openai.ChatCompletion.create(
model="gpt-4o",
messages=[{"role":"user","content":"Hello"}]
)
print(response.choices[0].message.content)
β οΈ Replace api_key with your key obtained after registering on FlowerWolf, and set base_url to our endpoint. Model names are the same as OpenAI's official ones β no code changes needed.
curl https://api.flowerwolf.net/v1/chat/completions \
-H "Content-Type: application/json" \
-H "Authorization: Bearer *** \
-d '{
"model": "gpt-4o",
"messages": [{"role":"user","content":"Hello"}]
}'
β οΈ Just replace your-flowerwolf-api-key. Supported models include gpt-4o, claude-3-5-sonnet, gemini-1.5-pro, etc.
import OpenAI from 'openai';
const client = new OpenAI({
apiKey: 'your-flowerwolf-api-key',
baseURL: 'https://api.flowerwolf.net/v1',
});
const response = await client.chat.completions.create({
model: 'gpt-4o',
messages: [{ role: 'user', content: 'Hello' }],
});
console.log(response.choices[0].message.content);
β οΈ Supports Node.js 18+. API is fully compatible with OpenAI β just change the URL and API key.
One API key to call all models
| Model | Type | Provider | Context | Use Cases |
|---|---|---|---|---|
| GPT-4o | Multimodal | OpenAI | 128K | All-purpose Assistant |
| GPT-4o-mini | Multimodal | OpenAI | 128K | Fast Response |
| Claude 3.5 SonnetPopular | Multimodal | Anthropic | 200K | Coding/Writing |
| Claude Opus 4 | Multimodal | Anthropic | 200K | Complex Reasoning |
| Gemini 1.5 Pro | Multimodal | 1M | Long Text Processing | |
| Gemini 1.5 Flash | Multimodal | 1M | Fast/Low Cost | |
| DeepSeek-V2.5 | Language | Domestic | 128K | Cost-effective Choice |
| Qwen2.5-72B | Language | Domestic | 128K | Chinese Scenarios |
More models being added continuously β full list available in the console after registration
Have your own GPU and servers? Want to run models yourself? We can help
Supports one-click deployment of OpenWebUI, Ollama, and other open-source UIs, with built-in model management and API endpoints.
Our proprietary AI Agent runtime framework, supporting multi-model switching, long-term memory, and tool calling.
Click "Login" in the top right, register with your email. You'll automatically receive trial credits β no top-up needed to try.
After logging in, go to "Console β API Keys" and copy your key. Format is similar to fw-xxxxxxxxxxxx.
Copy the example code above, replace your-flowerwolf-api-key with your actual key β just 2 lines of code.
Run the code β when you see the AI's response, it's connected. The console lets you check usage, view bills, and switch models.