Don't wait 18 months for a hardware rollout. Route your AI traffic through the StreetMP OS proxy by changing a single line of code.
The Real Cost of Waiting
Every day your AI traffic runs unguarded is a compliance liability. Legacy hardware vendors bill you for the privilege of waiting.
Procurement & RFP
3 monthsHardware Shipping
2 monthsData Centre Install
3 monthsAgent Rollout
6 monthsTraining & Tuning
4 monthsActive
Finally.Generate API Key
1 minUpdate Base URL
30 secAdd Tenant Header
30 secActive
Instantly.You save ~17 months, 29 days, and 22+ hours of deployment time.
That's 17 months of unguarded AI traffic your competitors don't have to worry about.
The Integration
Pick your language. Copy the snippet. You're protected.
from openai import OpenAIclient = OpenAI(api_key="sk-your-openai-key",base_url="https://api.openai.com/v1", # ← change this)
from openai import OpenAIclient = OpenAI(api_key="sk-your-openai-key",base_url="https://api.streetmp.com/v1/proxy", # ✅ donedefault_headers={"x-streetmp-key": "sk_live_••••••••••••••••",},)
Step by Step
Log in to the StreetMP OS dashboard. Under API Keys, click Generate. Your key is scoped to your tenant with full RBAC policy enforcement — ready in under 60 seconds.
In your existing OpenAI, Anthropic, or Google SDK client, change a single string: your base URL from `api.openai.com/v1` to `api.streetmp.com/v1/proxy`. Your code stays identical.
Pass your `x-streetmp-key` in the request headers. This binds every request to your Policy-as-Code rules, DLP engine, and Merkle audit trail.
Every prompt now passes through: Prompt Firewall → NeMo Guard () → PII Enclave Tokenisation → Merkle Audit. Your CISO gets a tamper-proof root hash on day one.
What You Get
Blocks jailbreaks and injection attempts before they reach any AI model.
NVIDIA NeMo secondary deep evaluation. Fail-open so no downtime.
Every request generates a cryptographic leaf. One root hash for your CISO.
Your raw data never reaches the AI provider. Tokenised before dispatch.
Mathematical proof of policy compliance attached to every response.
OpenAI, Anthropic, Google, Mistral, or your own local LLM — zero lock-in.
Your API key is ready the moment you sign up. No sales call required to start. Full enterprise contracts available for SOC2-regulated environments.