集成我们的统一 AI API 网关所需的一切。 通过全面的指南和示例，几分钟内即可开始使用。
17 of 25 topics completed
Start making API calls immediately. Access 400+ AI models with one unified interface.
Try API# Install the SDK
pip install barq
# Initialize client
from barq import AIClient
client = AIClient(api_key="your_api_key")
# Make your first request
response = client.chat.completions.create(
model="gpt-4",
messages=[{"role": "user", "content": "Hello!"}]
)
print(response.choices[0].message.content)Core endpoints and authentication
All API requests require an API key in the Authorization header.
Authorization: Bearer your_api_key/v1/chat/completionsCreate a chat completionAuth required/v1/embeddingsCreate embeddingsAuth required/v1/modelsList available modelsAuth required/v1/usageGet usage statisticsAuth required15+ regions worldwide with <100ms latency
Access to all major providers in one API
SOC 2 compliant with end-to-end encryption
Optimized routing and automatic failover
Real-time usage monitoring and cost optimization
We offer a free tier with $10 in credits. Paid plans start at $29/month with higher limits and advanced features. See our pricing page for details.
We support 400+ models including GPT-4, Claude 3, Gemini Pro, Llama 3, and many more. New models are added weekly.
Yes. We are SOC 2 compliant, use end-to-end encryption, and have a strict zero data retention policy. Your data never leaves our secure infrastructure.
Our platform automatically routes requests to alternative models with similar capabilities, ensuring 99.9% uptime.
Our support team is available 24/7 to help you with any questions or issues.