One API. Every possibility.Build more. Spend less.
Leading AI models. One integration. Transparent, usage-based pricing.
From your first request to model routing and cloud infrastructure.
OpenAI SDK compatible · No credit card required
from openai import OpenAI
client = OpenAI(
base_url="https://api.promptoll.com/v1",
api_key="zai_•••"
)
client.chat.completions.create(
model="auto",
messages=[{"role": "user", "content": "Hello"}]
)Keep your existing SDK
Track usage and cost per request
Manage compute in the same console
Model routes
—
across 11 providers — OpenAI, Anthropic, Google, DeepSeek, Qwen, GLM, Kimi, MiniMax, Doubao, and more
Compute integration
Available
compute launch, lifecycle, and status dashboards
API compatibility
OpenAI
drop-in SDK calls are supported
From integration to production.
The OpenAI API on the outside. A cost optimization engine on the inside.
OpenAI + Anthropic compatible
Drop-in replacement for both the OpenAI SDK and the Anthropic Messages API (tool use, vision, extended thinking included) — change the base_url and you're done.
Multi-tenant keys
Issue scoped zai_ keys, set per-tenant markup, revoke instantly. Ideal for reselling AI inside your product.
Leading models. Find your fit.
From everyday tasks to complex reasoning. Compare prices, then choose your model.
| Model | Input / 1M tokens | Output / 1M tokens |
|---|---|---|
USD prices from the current catalog. See the pricing page for full billing details.
Beyond models. Your infrastructure, too.
VPC isolation, snapshots, custom images, independent data disks, elastic bandwidth — the same depth you'd expect from a dedicated cloud provider, running underneath your AI gateway.
See compute capabilitiesPer-tenant VPC isolation
Every tenant gets its own isolated VPC and subnet, provisioned automatically on first order — no shared network surface with other tenants.
Snapshots & custom images
Snapshot a disk or instance at any point, or capture a stopped instance as a reusable custom image to launch fresh copies from.
Flexible data disks
Create disks at any size, resize later, attach or detach while the instance is stopped — provisioning that scales independently of compute.
On-the-fly EIP bandwidth
Adjust your elastic IP's public bandwidth up or down whenever traffic changes, billed prorated to the minute.
Full visibility into every request
Every request shows requested vs. actual model, token usage, and what you were charged. Audit the full arbitrage effect — no hand-waving.
Price illustration: USD / 1M tokens (input / output), blended at 1:3. Not an actual routing result or guaranteed savings.
{
"model_requested": "gpt-4o",
"model_used": "deepseek-v4-pro",
"prompt_tokens": 412,
"completion_tokens": 189,
"cost_microcents": 23940,
"price_microcents": 86450,
"latency_ms": 612,
"stream": true
}Four steps to your first request.
From creating a key to reviewing logs, every step has a clear starting point.
Sign up
Create an account — no credit card. Get a zai_ API key in seconds.
Swap one line
Point your SDK base_url to api.promptoll.com/v1 and keep your existing calls.
Save instantly
Use model="auto" or opt into arbitrage. Watch cost drop in the dashboard.
Use in production
Call it from production traffic and monitor with traceable logs and host state.
It starts with one request.
Start free. Pay only for what you forward. Cancel any time.