chuangyeshuo/aiephant-ai-gateway

Set these on your host before deploying. Each platform has an “Environment Variables” section in the deploy flow. (showing first 20, more exist in .env.template)

• ALEPHANT_CONTROL_PLANE_API_KEY secret required if you want llm observability, authentication
• OPENAI_API_KEY secret provider keys, only set the ones you wish to enable
• ANTHROPIC_API_KEY secret
• GEMINI_API_KEY secret
• AWS_ACCESS_KEY secret
• AWS_SECRET_KEY secret
• OTEL_METRIC_EXPORT_INTERVAL
• AI_GATEWAY__DEPLOYMENT_TARGET__TYPE we use cloud mode
• POSTGRES_DATABASE_URL secret database url
• AI_GATEWAY__ALEPHANT__FEATURES Compatibility mode: start without PostgreSQL; also set `AI_GATEWAY__ALEPHANT__FEATURES=none` (no auth, no Alephant request-log forwarding).…
• AI_GATEWAY__ALEPHANT__BASE_URL
• S3_ENDPOINT Object storage (S3-compatible API): maps to YAML `s3`, then AI_GATEWAY__S3 , then flat S3_* overrides (S3_* wins when set and non-empty).
• S3_ACCESS_KEY secret
• S3_SECRET_KEY secret
• S3_BUCKET_NAME
• S3_REGION
• MASTER_KEY_ENCRYPTION_KEY secret Optional: path | virtual-host, same as YAML `url-style` S3_URL_STYLE=path master key encryption key
• AI_GATEWAY__REQUEST_LOG__TRANSPORT Request-log Redis Stream; same URL is used for workspace concurrency and model-catalog EXISTS. If URL is unset with transport=redis, falls…
• REDIS_URL secret
• REDIS_STREAM_KEY_REQUEST_RESPONSE Override Stream key when aligned with logs-collector deployment: