OpenAI-compatible API for 20+ leading models

BoostRail exposes an OpenAI-compatible API — the same /v1/chat/completions and /v1/models your OpenAI SDK already calls. Point base_url at the gateway, keep your code, and reach 20+ models from 8 providers through one key.

What “OpenAI-compatible” means

An OpenAI-compatible API speaks the same request and response shape as OpenAI's own API: the same endpoints (/v1/chat/completions, /v1/models), the same JSON fields, the same streaming format, and the same Authorization: Bearer header. A client built for the OpenAI SDK works against it unchanged.

So you don't adopt a new SDK or rewrite your integration. You change two values — base_url and api_key — and every request keeps working, now routed through BoostRail to any model in the catalog.

Why call models through one OpenAI-compatible endpoint

One integration, every model: Claude, GPT, Gemini, DeepSeek, GLM, Grok, Kimi and MiniMax behind a single API key — no separate account, SDK, or bill per provider.
Switch models by changing one string: the model parameter picks the model; nothing else in your code changes.
Streaming behaves the same: set stream=true and responses arrive in the standard OpenAI server-sent-events format.
Per-token billing in USD: one prepaid balance, transparent per-token prices, one statement across every model.
Bring your own keys (BYOK): route through your own provider keys when you have them, with the platform as automatic fallback — plus 1,000,000 free BYOK requests every month.

How to use it

If your app already uses the OpenAI SDK, this is the whole change — swap base_url and api_key, keep the rest:

from openai import OpenAI

client = OpenAI(
    api_key="YOUR_API_KEY",
    base_url="https://api.boostrail.com/v1",
)
response = client.chat.completions.create(
    model="claude-sonnet-5",
    messages=[{"role": "user", "content": "Hello"}],
)
print(response.choices[0].message.content)

Models you can reach

One OpenAI-compatible endpoint reaches Anthropic Claude, OpenAI GPT, Google Gemini, DeepSeek, GLM (Zhipu), xAI Grok, Kimi (Moonshot) and MiniMax — 20+ models with public per-token USD pricing on the Models page, growing as providers are onboarded.