OpenAI-compatible gateway

OpenAI-Compatible API — GPT, Claude & Gemini on One Endpoint

LumeAPI exposes GPT, Claude, Gemini, image, and video models through a single OpenAI-compatible base URL — change your API key, base URL, and model id without rewriting application code.

Works with the OpenAI Python/Node SDKs, curl, Cursor, LangChain, and most agents that already call /v1/chat/completions.

Supported models on the compatible endpoint

Official reference vs LumeAPI catalog rates. Pricing unit: per 1M input / output tokens. Last updated: July 2026. Source: provider list price.

ModelOfficial (in / out)LumeAPI (in / out)Savings
GPT-5.6 Terragpt-5.6-terra$2.50 / $15.00$1.25 / $7.5050% offDetails →
Claude Sonnet 4.6claude-sonnet-4-6$3.00 / $15.00$1.80 / $9.0040% offDetails →
Gemini 3.5 Flashgemini-3.5-flash$1.50 / $9.00$0.90 / $5.4040% offDetails →

Monthly cost examples

Illustrative totals for gpt-5.6-terra using catalog list prices — your actual bill depends on retries, tool loops, and output length.

SDK migration

15M input + 4M output tokens / month

Existing OpenAI app with moderate traffic

Official (GPT-5.6 Terra)
$97.50/mo
LumeAPI
$48.75/mo
Monthly savings
$48.7550% off

Rates last updated July 2026

Multi-provider product

60M input + 18M output tokens / month

GPT for chat, Claude for analysis, Gemini for RAG

Official (GPT-5.6 Terra)
$420.00/mo
LumeAPI
$210.00/mo
Monthly savings
$210.0050% off

Rates last updated July 2026

Migrate in minutes

Three values change: API key, base URL, model id. Everything else stays the same.

Before (official OpenAI)

python
from openai import OpenAI

client = OpenAI(api_key="YOUR_OPENAI_API_KEY")

response = client.chat.completions.create(
    model="gpt-5.6-terra",
    messages=[{"role": "user", "content": "Hello"}],
)

After (LumeAPI)

python
from openai import OpenAI

client = OpenAI(
    api_key="YOUR_LUMEAPI_KEY",
    base_url="https://api.lumeapi.site/v1",
)

response = client.chat.completions.create(
    model="gpt-5.6-terra",
    messages=[{"role": "user", "content": "Hello"}],
)

Full step-by-step rollout, streaming checks, and FAQ: OpenAI API Too Expensive? migration guide →

Trust & billing

Is the API shape identical to OpenAI?

Chat Completions, image generations, and async video follow OpenAI-compatible paths documented in /docs. Provider-specific features may differ — test your payload.

Which models can I call?

Every text model in our catalog uses POST /v1/chat/completions. Image and video models use their documented endpoints — see per-model docs.

How do I authenticate?

Authorization: Bearer YOUR_LUMEAPI_KEY on every request. Keys are created in Console after registration.

Is my data used to train models?

See Privacy Policy for retention and subprocessors. You are responsible for what you send through the gateway.

Built for these workloads

Drop-in base URL swap

Keep Chat Completions request bodies. Set base_url to https://api.lumeapi.site/v1 and swap model id to a catalog entry — no new SDK required.

Cursor and IDE assistants

Configure custom OpenAI endpoint in Cursor settings. Use LumeAPI key and catalog model ids for lower-cost daily coding traffic.

LangChain & agent frameworks

Any framework that accepts openai_api_base works out of the box. Route different chains to GPT, Claude, or Gemini by model id only.

Staging before production

Mirror production traffic on LumeAPI in a staging environment. Compare latency, streaming, and tool-call behavior per model.

Related guides

FAQ

Do I need the Anthropic or Google SDK?

No for standard chat. Use the OpenAI SDK with LumeAPI base URL and catalog Claude or Gemini ids. Test tool calling for your agent use case.

What about embeddings or Assistants API?

LumeAPI focuses on Chat Completions and multimodal generation documented in /docs. Check catalog for supported endpoints before migrating niche APIs.

Does streaming work?

Yes on supported models with stream: true. Validate reconnect and partial JSON handling in your client before production.

Can I use the same key for image and video?

Yes. One wallet debits all modalities. See model pages for image and video parameter differences.

How fast is migration?

Most teams change three env vars and redeploy. Allow time to regression-test streaming, tools, and error handling on top models.

Where is the full API reference?

Start at /docs for gateway overview, then /docs/models/{id} for per-model parameters and curl examples.

Point your OpenAI client at LumeAPI

Register, create a key, and set base_url to https://api.lumeapi.site/v1. Need help choosing a model? Browse the developer docs or contact support.