ZevRouter
Get started
☰

Agent setup

Use ZevRouter from your AI tools

Configure ZevRouter as an OpenAI-compatible provider with a base URL, model ID, and API key.

Setup Quickstart

Provider Config

Base URL: https://zusd-testnet---zevrouter-staging-skwx3z2gxa-uc.a.run.app/api/v1
Model: google/gemini-3.8-flash
API key: zvr_REPLACE_ME

Environment Variables

export OPENAI_BASE_URL="https://zusd-testnet---zevrouter-staging-skwx3z2gxa-uc.a.run.app/api/v1"
export OPENAI_API_KEY="zvr_REPLACE_ME"
export OPENAI_MODEL="google/gemini-3.8-flash"

Key Safety

Verification Request

curl "$OPENAI_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3.8-flash",
    "max_tokens": 64,
    "messages": [
      { "role": "user", "content": "Reply with: ZevRouter setup OK" }
    ]
  }'

Optional Routing Check

curl "$OPENAI_BASE_URL/chat/completions" \
  -H "Authorization: Bearer $OPENAI_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "google/gemini-3.8-flash",
    "routing": "greenest",
    "max_tokens": 64,
    "messages": [
      { "role": "user", "content": "Reply with one short clean energy sentence." }
    ]
  }'

Usage Check

curl "$OPENAI_BASE_URL/me" \
  -H "Authorization: Bearer $OPENAI_API_KEY"

Prompt To Give The User's Agent

Configure my local AI harness to use ZevRouter as an OpenAI-compatible provider.

Use:
Base URL: https://zusd-testnet---zevrouter-staging-skwx3z2gxa-uc.a.run.app/api/v1
Model: google/gemini-3.8-flash
API key: I will provide it separately as a secret.

Rules:
Do not print the full API key.
Do not commit the key to git.
Prefer env vars, keychain, encrypted config, or an ignored local .env file.
Inspect the existing harness config before editing.
After setup, run one minimal chat completion and one /api/v1/me usage check.
Report whether it worked, which model was used, and the request usage.

Common Errors