Helicone is a platform that enables you to build dependable AI applications by providing an AI gateway along with observability tools for monitoring and optimizing LLM usage.
1.1 API Configuration
Method 1: Helicone AI Gateway
Helicone can forward your requests to your inference API using the Helicone-Target-URL header:
import openai
client = openai.OpenAI(
api_key="YOUR_RYGEN_API_KEY",
base_url="https://ai-gateway.helicone.ai/v1",
default_headers={
"Helicone-Auth": "Bearer YOUR_HELICONE_API_KEY",
"Helicone-Target-URL": "https://api.rygen.io/v1"
}
)
response = client.chat.completions.create(
model="your-model-name",
messages=[{"role": "user", "content": "Hi there!"}]
)Method 2: Self-Hosted Helicone
export RYGEN_API_KEY="sk-your-rygen-api-key"
export HELICONE_API_KEY="sk-helicone..."
curl --request POST \
--url http://localhost:8585/v1/gateway/oai/v1/chat/completions \
--header "Authorization: Bearer $RYGEN_API_KEY" \
--header "Helicone-Auth: Bearer $HELICONE_API_KEY" \
--header "Helicone-Target-URL: https://api.rygen.io/v1" \
--header "Content-Type: application/json" \
--data '{
"model": "your-model-name",
"messages": [{"role": "user", "content": "Hi, How are you doing?"}]
}'