For developers

The Lobstack API.

OpenAI-compatible. Change one base URL, and every response tells you which model answered and what it cost.

Change one line.

Keep your OpenAI SDK. Point it at the Gateway with a Lobstack key.

from openai import OpenAI

client = OpenAI(
    base_url="https://www.lobstack.ai/api/gateway/v1",
    api_key="lsk_live_…",
)

client.chat.completions.create(model="auto", messages=[…])

Every call has a receipt.

The model that answered, the tokens, the cost, and what the same tokens would have cost on the priciest model your plan allows.

How metering works →

Sample· Developer plan

You asked forauto

The router servedClaude Haiku 4.5

Tokens412 in · 168 out

Cost$0.001565

The same tokens on Claude Fable 5.1, the priciest model the plan allows$0.0157

Saved on this request$0.0141

Nex picks the model.

Send model: "auto" and Nex, the router, picks a model for each request, within your plan.

Run it on your own prompt →How Nex decides →

01Your request is scored for how hard it is.
02The cheapest tier that can answer it serves it.
03Never above the ceiling your plan sets.

See every call, key and client in Console.

Console is your account: usage, logs, keys, clients and billing. Log in to open it.

Console: usage, spend by client, request logs and API keys, on sample data

Recorded on sample data.

See which models it serves, and the rate for each →