A single, OpenAI-compatible API to frontier and open-source models. Fund a prepaid balance with a gift card and start building.
gc4.ai is a unified inference API for frontier and open-source language models, billed from a
prepaid balance you fund with a gift card. The surface is OpenAI-compatible, so the tools you
already use work unchanged — point your base URL at https://gc4.ai/v1 and go.
- OpenAI-compatible —
/v1/chat/completions,/v1/completions,/v1/models. Drop-in for the OpenAI SDKs. - Frontier + open models — Anthropic, Google, OpenAI (gpt-oss), Meta, Qwen, DeepSeek, and more, through one endpoint.
- Prepaid — fund your balance with a gift card; pay only for what you use.
- Streaming — Server-Sent Events with token-usage accounting.
Use any OpenAI-compatible client — just change the base URL:
curl https://gc4.ai/v1/chat/completions \
-H "Authorization: Bearer $GC4AI_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "google/gemini-2.5-flash",
"messages": [{ "role": "user", "content": "Hello!" }]
}'Prefer a typed client? Grab one of the SDKs below.
Every SDK is generated from one OpenAPI 3.1 contract, so they stay in lockstep with the live API.
| Repository | What it is |
|---|---|
| typescript-sdk | Type-safe TypeScript SDK + MCP server, with typed SSE streaming. |
| python-sdk | Python SDK for the gc4.ai API. |
| openapi | The OpenAPI 3.1 contract — the source of truth behind every SDK. |
Built by PRVN Labs, LLC · gc4.ai