Skip to main content

​
What is the OpenAI Compatible Interface

OpenAI’s Chat Completions API format has become the industry de facto standard. Many large language models provide OpenAI-compatible interfaces, allowing developers to access different models using the same code. Ace Data Cloud offers OpenAI-compatible Chat Completions interfaces for models such as Claude, Gemini, DeepSeek, Grok, Kimi, etc. You can switch models simply by changing the Base URL and model name.

​
Supported Models

ServiceEndpointExample Model
ClaudePOST /v1/chat/completionsclaude-sonnet-4-6
OpenAIPOST /openai/chat/completionsgpt-4o
GeminiPOST /gemini/chat/completionsgemini-2.5-flash
DeepSeekPOST /deepseek/chat/completionsdeepseek-r1
GrokPOST /grok/chat/completionsgrok-3
KimiPOST /kimi/chat/completionskimi-k2.5

​
Unified Invocation Method

All chat models use the same request format: 
import requests

def chat(endpoint, model, message):
    return requests.post(
        f"https://api.acedata.cloud{endpoint}",
        headers={
            "Authorization": "Bearer YOUR_API_TOKEN",
            "Content-Type": "application/json",
        },
        json={
            "model": model,
            "messages": [{"role": "user", "content": message}],
            "max_tokens": 1024,
            "temperature": 0.7,
        },
    ).json()

# Calling different models - just change the endpoint and model name
claude = chat("/v1/chat/completions", "claude-sonnet-4-6", "你好")
gpt = chat("/openai/chat/completions", "gpt-4o", "你好")
gemini = chat("/gemini/chat/completions", "gemini-2.5-flash", "你好")
deepseek = chat("/deepseek/chat/completions", "deepseek-r1", "你好")

​
Streaming Output

All chat APIs support streaming output (Server-Sent Events): 
data = {
    "model": "claude-sonnet-4-6",
    "messages": [{"role": "user", "content": "写一首诗"}],
    "stream": True,
}

response = requests.post(
    "https://api.acedata.cloud/v1/chat/completions",
    headers={"Authorization": "Bearer YOUR_API_TOKEN"},
    json=data,
    stream=True,
)

for line in response.iter_lines():
    if line:
        print(line.decode())

​
Compatibility with OpenAI SDK

Since the interface format is compatible, you can directly use the official OpenAI SDK by just modifying the base_url: 
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_TOKEN",
    base_url="https://api.acedata.cloud/v1",  # Claude
    # base_url="https://api.acedata.cloud/openai",  # OpenAI
    # base_url="https://api.acedata.cloud/gemini",  # Gemini
)

response = client.chat.completions.create(
    model="claude-sonnet-4-6",
    messages=[{"role": "user", "content": "你好"}],
)
print(response.choices[0].message.content)

​
Summary

With Ace Data Cloud’s OpenAI compatible interface, you can:
  • Access 6+ large language models with a single codebase
  • Flexibly switch models to compare performance and cost
  • Use the OpenAI SDK directly with zero migration cost
  • Unified billing and monitoring
Visit platform.acedata.cloud for a free trial.