Instead of juggling accounts and SDKs, you point one OpenAI-compatible key at TDSPRO and reach 20+ models. Here is how to choose the right one without memorizing a catalog.
Capability aliases
The simplest way to choose is by what you need done. Pass one of these as model and smart routing picks a strong fit:
fast
Low-latency answers, autocomplete, tagging at scale.
smart
Hard reasoning, architecture, careful long answers.
code
Writing and fixing code, reviewing diffs.
vision
Reading images, screenshots, and charts.
chat
Friendly back-and-forth, drafting, brainstorming.
// Let routing choose a great model for the job:
await client.chat.completions.create({
model: "code",
messages: [{ role: "user", content: "Refactor this function for readability." }],
});Or call a model by name
Need a specific flagship like GPT-4.1? Pass the exact name. The response always reports the real model — no silent substitutions.
await client.chat.completions.create({
model: "gpt-4.1",
messages: [{ role: "user", content: "Draft a product announcement." }],
});Want proof? Every reply carries the model field with the true name. In our Chat and Arena we show that exact name on each message — compare it against the official API any time.
Switching is one line
Because everything is OpenAI-compatible, moving between models — or from a direct integration to TDSPRO — is a base-URL and a string change. No rewrites.
Get your one key
Create a free key and reach all 20+ models in a minute. No card to start.