Fast and affordable multimodal model for lightweight tasks.
Provider costs passed through at-rate. Zero markup. All inference encrypted in hardware.
Drop-in replacement for the OpenAI SDK. Point your base URL to Zima and every request is hardware-encrypted automatically.
1import OpenAI from "openai"
2
3const client = new OpenAI({
4 apiKey: process.env.ZIMA_KEY,
5 baseURL: "https://api.zima.ai/v1",
6})
7
8const response = await client.chat.completions.create({
9 model: "gpt-4o-mini",
10 messages: [{ role: "user", content: "Hello" }],
11})1from openai import OpenAI
2
3client = OpenAI(
4 api_key=os.environ["ZIMA_KEY"],
5 base_url="https://api.zima.ai/v1",
6)
7
8response = client.chat.completions.create(
9 model="gpt-4o-mini",
10 messages=[{"role": "user", "content": "Hello"}],
11)