Install
Drop in the OpenAI SDK. No proprietary client, no vendor lock. One dependency you already know.
Hardware encrypted inference.
Zero logs. Zero limits.
An encrypted AI gateway for every model.
Inference leading LLMs run inside verifiably secure runtimes, powered by Intel TDX and NVIDIA Confidential Computing architectures.
1import OpenAI from "openai"
2
3const zima = new OpenAI({
4 baseURL: "https://zimalabs.io/api/v1",
5 apiKey: process.env.ZIMA_KEY,
6})
7
8const response = await zima.chat.completions.create({
9 model: "deepseek-v3.1-terminus",
10 messages: [{ role: "user", content: "Explain confidential computing in one sentence." }],
11})
12
13// encrypted in hardware. zero data retained.Seamless migration to secure compute
Use your existing OpenAI client. Swap the base URL and API key to start encrypting every inference request today.
Point to Zima
Swap the base URL. Every request now routes through our secure enclave — encrypted in hardware, zero data retained.
Ship to Production
Same models, same latency, hardware-level encryption. Your compliance team will thank you.
From SDK request to protected output, without storing prompts or responses.
Secure by design
From key exchange to model output, every step runs inside hardware-attested enclaves.
Provider costs passed through at-rate. Zero markup.
| Model | Input / 1M | Output / 1M |
|---|---|---|
| GPT-4o | $2.50 | $10.00 |
| Claude Sonnet 4 | $3.00 | $15.00 |
| Gemini 2.5 Pro | $1.25 | $5.00 |
| Llama 3.1 70B | $0.50 | $0.75 |
| Mistral Large | $2.00 | $6.00 |
Secure Enclave
Zero bytes retained
Every request executes inside a hardware-isolated enclave. No prompts, outputs, or metadata survive past completion.
Security, compatibility, and pricing.
No. Requests exist only for the duration of execution inside the enclave. No prompts, outputs, metadata, or logs are persisted. This is enforced by hardware, not policy.
Access is blocked immediately. If the system cannot cryptographically verify that the runtime environment is running approved, unmodified software, no data is processed.
Yes. Zima uses the chat/completions format. Point your existing OpenAI SDK at our base URL and it works — same request shape, same response shape, zero retention.
Every model in the catalog is labeled with its security tier. TEE-hosted models run entirely inside hardware enclaves.
Per-token, per-model, pay-as-you-go. No platform fees, no hidden tiers, no commitments. Pricing is displayed on every model card before you send a request.
$10 in free credits. Start encrypting your inference today.
Point the OpenAI SDK at Zima, keep model choice and pricing visibility, and move sensitive traffic into attested hardware.