OpenAI to GonkaGate Migration Guide

Move an existing OpenAI-compatible chat.completions integration to GonkaGate by changing the base URL, API key, and model ID, then verify the switch with GET /v1/models and one POST /v1/chat/completions request. Keep the migration narrow: change transport first, prove parity, then touch prompts or broader app logic.

Before you switch

A verified GonkaGate account
A saved gp-... API key
A fresh model ID from GET /v1/models for the workload you are migrating
Enough prepaid USD balance for the smoke test and rollout
A working OpenAI-compatible path in your app

Change only three values

Set base_url or baseURL to https://api.gonkagate.com/v1
Replace the OpenAI key in sk-... format with a GonkaGate key in gp-... format
Replace the OpenAI model name with a fresh GonkaGate model ID from GET /v1/models

Request Example

 from openai import OpenAI

client = OpenAI(
-    base_url="https://api.openai.com/v1",
+    base_url="https://api.gonkagate.com/v1",
-    api_key="sk-..."
+    api_key="gp-..."
)

response = client.chat.completions.create(
-    model="gpt-5.2",
+    model="qwen/qwen3-235b-a22b-instruct-2507-fp8",
     messages=[{"role": "user", "content": "Say hello from GonkaGate"}]
)

Verify the migration

curl https://api.gonkagate.com/v1/models \
  -H "Authorization: Bearer gp-your-api-key"

Command

curl https://api.gonkagate.com/v1/chat/completions \
  -H "Authorization: Bearer gp-your-api-key" \
  -H "Content-Type: application/json" \
  -d '{
    "model": "qwen/qwen3-235b-a22b-instruct-2507-fp8",
    "messages": [
      { "role": "user", "content": "Say hello from GonkaGate" }
    ]
  }'

Expected result:

GET /v1/models returns a model list
POST /v1/chat/completions returns a normal response with choices[0].message.content

Common migration failures

Response	What it usually means	What to do
`401 invalid_api_key`	The Bearer header, key value, or account state is wrong	Recheck the `Authorization: Bearer gp-...` header and confirm the key is active
`404 model_not_found`	The model ID is stale or unsupported	Refresh the model ID from `GET /v1/models` before treating the migration as broken
`429 insufficient_quota`	The account does not have enough available balance	Add balance or lower test traffic before rollout
`429 rate_limit_exceeded`	You hit a runtime limit	Handle it with bounded retries instead of treating it as a migration-only failure
`5xx`	A temporary upstream or platform error occurred	Retry with bounded backoff and confirm whether the failure persists

Compatibility notes

usage can include additive cost fields such as base_cost_usd, platform_fee_usd, and total_cost_usd, so strict response typing should allow extra fields.
This guide covers chat.completions and GET /v1/models. If your app depends on embeddings, Responses API, Assistants, Audio, Batch, or fine-tuning, keep those flows on a different provider path for now.
Streaming, tools, JSON mode, and vision depend on the selected model and API surface. Confirm support for the exact model before rollout.

OpenAI to GonkaGate Migration

Before you switch

Change only three values

Verify the migration

Common migration failures

Compatibility notes

See also

On this page