Enterprise AI API Gateway
Managed OpenAI access for your organisation. Drop-in API proxy with token-based billing, usage dashboards, and admin controls.
Why Use an AI API Gateway?
Giving every team direct access to OpenAI API keys creates billing chaos, zero visibility into usage, and no way to control costs. An API gateway centralises access — one key per team, tracked usage, and admin-controlled budgets.
Cost Control
Pre-allocate token budgets per team. No surprise bills from runaway scripts.
Usage Visibility
Track every request — which team, which model, how many tokens, when.
Key Security
One master key stays with IT. Teams get scoped Ogma tokens — revocable instantly.
Model Governance
Control which models are available. Block expensive models for non-production use.
How It Works
1. Get API Key
Sign up on the Ogma portal, create an API token. Admin loads token credits to your account.
2. Send Requests
Point your OpenAI SDK to our endpoint. Same models, same parameters, same response format.
3. Track & Control
View real-time usage dashboards. Admin sets budgets, controls models, manages credits.
Drop-In Replacement
Already using the OpenAI SDK? Change two lines — base_url and api_key. Everything else stays identical. Models, parameters, streaming, function calling — all work out of the box.
- OpenAI-compatible /v1/chat/completions endpoint
- Streaming (SSE) support
- All GPT-4.1 and GPT-4o models
- Balance check via /api/balance
- Model listing via /v1/models
from openai import OpenAI
client = OpenAI(
base_url="https://ogma.in/apps/ai-gateway/v1",
api_key="OgmaAI-YOUR-TOKEN-HERE",
)
response = client.chat.completions.create(
model="gpt-4.1",
messages=[{"role": "user", "content": "Summarise this report..."}],
)
print(response.choices[0].message.content)
curl https://ogma.in/apps/ai-gateway/v1/chat/completions \
-H "X-API-Token: OgmaAI-YOUR-TOKEN" \
-H "Content-Type: application/json" \
-d '{"model":"gpt-4.1","messages":[{"role":"user","content":"Hello"}]}'
Platform Features
Real-Time Dashboard
Balance, usage charts, model breakdown, and recent requests — all in one view.
Token Management
Create, name, and revoke API tokens instantly. Each token is scoped to one user.
Credit Ledger
Full audit trail of every credit and debit. Know exactly where tokens went.
Model Pricing
Admin sets per-model multipliers. Charge more for premium models, less for nano.
Multi-User
Each team or developer gets their own tokens and balance. Admin manages centrally.
Enterprise Security
SHA-256 hashed tokens, HTTPS only, no API keys exposed to end users.
Token-Based Pricing
Pay only for what you use. No monthly minimums, no per-seat charges.
Starter
For small teams and proof-of-concept
loaded on signup
- All models
- Portal dashboard
- Streaming support
Enterprise
For production workloads
volume-based pricing
- Volume discounts
- Dedicated support
- SLA guarantee
- Custom model config
Frequently Asked Questions
Ready to Get Started?
Get your API key, integrate in 5 minutes, and start using GPT-4.1 with full cost control and visibility.