Run the EveryAPI gateway inside Helicone

One base URL line points Helicone at EveryAPI

Default global Base URL / China-first Base URL: https://api.everyapi.ai/v1 / https://api-cn.everyapi.ai/v1

# Helicone proxy mode: change the base URL to oai.helicone.ai/v1,
# and use the Helicone-Target-Url header to specify the real upstream (EveryAPI)
 
from openai import OpenAI
 
client = OpenAI(
    base_url="https://oai.helicone.ai/v1",
    api_key="sk-everyapi-...",                  # EveryAPI key
    default_headers={
        "Helicone-Auth": "Bearer sk-helicone-...",         # Helicone key
        "Helicone-Target-Url": "https://api.everyapi.ai",  # Forward target
    },
)
 
resp = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[[{"role": "user", "content": "Hello"}]],
)

From zero to your first call

1. 01
   Sign up for Helicone

   Create or self-host a Helicone project using its current documentation and obtain the credentials required by the integration mode you choose.

   us.helicone.ai

2. 02
   Decide between proxy and async logger

   Proxy mode changes the base URL to oai.helicone.ai/v1 and points Helicone-Target-Url at EveryAPI. With an async logger, requests go to EveryAPI directly and telemetry is submitted separately. Each mode has different networking, privacy, and failure trade-offs; validate it in your environment.

3. 03
   How the API keys relate

   In proxy mode, keep the EveryAPI key as the OpenAI client credential and provide the Helicone credential through the header documented by Helicone. Confirm the current target-URL format in Helicone’s docs before deployment.

4. 04
   View traces in the Helicone dashboard

   Confirm which fields your instrumentation sends to Helicone. Treat the EveryAPI dashboard as authoritative for EveryAPI usage and billing.

EveryAPI plugs into these too