Linkrra Log in Sign up free

Guide · Portkey AI Gateway

Connect Portkey AI Gateway to Synap

3 steps · about 2 minutes

  1. Get both API keys

    Create a free Synap key at synap402.com, then copy a PORTKEY_API_KEY from your Portkey workspace.

  2. Route one request through Portkey

    Portkey authenticates with its own header. The regular Authorization header is forwarded to Synap, and x-portkey-custom-host selects our OpenAI-compatible base URL.

    curl https://api.portkey.ai/v1/chat/completions \
  -H "x-portkey-api-key: $PORTKEY_API_KEY" \
  -H 'x-portkey-provider: openai' \
  -H 'x-portkey-custom-host: https://synap402.com/v1' \
  -H "Authorization: Bearer $SYNAP_API_KEY" \
  -H 'content-type: application/json' \
  -d '{"model":"qwen3-coder-30b","messages":[{"role":"user","content":"Say OK"}]}'

  3. Save it as an integration

    In Model Catalog → Integrations, add a Local / privately hosted provider. Choose the OpenAI API spec and use the same host and authorization header.

    Provider type   Local / privately hosted
API spec        OpenAI
Custom host     https://synap402.com/v1
Authorization   Bearer $SYNAP_API_KEY
Notes, prices and limits

A custom host and the standard OpenAI request shape. Recommended model: qwen3-coder-30b — $0.0675 in and $0.27 out per 1M tokens, 98,304-token context.

Portkey uses two credentials on this route: x-portkey-api-key authenticates the gateway, while Authorization: Bearer … is forwarded upstream and authenticates Synap.

The custom host includes /v1: https://synap402.com/v1. The request path shown by Portkey is then /chat/completions, so the upstream request lands on the standard Synap endpoint.

Portkey logging, tracing and retention are controlled by your Portkey workspace. Synap does not retain prompts or completions after the request reaches us.

The header names and integration fields were checked against Portkey’s official custom-provider documentation; the Synap model catalogue and upstream endpoint were checked live.

On the free tier: every request counts input and output against 100,000 free tokens a day, and answers stop at 4,096 output tokens. Coding agents send long prompts, so the free tier covers a short session; credits cover the rest.

Field names from Portkey docs: custom hosts and self-hosted providers, checked 2026-11-10 10:24 UTC. Prices refresh from /v1/models when this page loads.