A queue that never shrinks
Password resets, access requests and vague 'it's broken' tickets bury the work that matters.
For IT and service desks
Clear the queue before standup
Triage tickets, summarise noisy logs, draft runbooks and answer the same how-do-I question for the hundredth time. One endpoint your team controls, with every call on record.
Faster triage
tickets categorised and routed on arrival
One governed endpoint
instead of shadow AI tools
Full audit trail
every request in history
The problem
What is slowing you down today
Shadow AI everywhere
Staff paste things into whatever tool they found. You have no keys, no limits and no visibility.
Logs nobody has time to read
The answer is in there somewhere, between thousands of lines of noise.
Capabilities
Everything you need, nothing you don't
Ticket triage
Classify category, urgency and team from the ticket text, returned as JSON your service desk can act on.
Keys you can govern
Keys are shown once and stored only as a hash. Create one per tool and revoke any of them in one click.
Rate limits per account
Set a requests-per-minute ceiling so one script cannot flood the endpoint.
Request history
Every call is recorded with model, tokens, latency and price, so you can see who is using what.
Code-capable routing
Ask for capability code when you need a script, a query or a config snippet explained.
Fail-over built in
Your automations keep running when an individual model is busy.
Use cases
What teams run on it
01
Service desk triage
Auto-tag incoming tickets and draft a first response for the agent to approve.
02
Log and alert summaries
Paste the noisy excerpt, get the likely cause and the next thing to check.
03
Runbook drafting
Turn a resolved incident thread into a clean, step-by-step runbook.
04
Internal knowledge answers
Answer how-to questions from your own documentation, passed in the prompt.
How it works
From sign-up to production
- 1
Create named keys
One key per integration, so you can revoke any tool's access without touching the rest.
- 2
Set your guardrails
Choose a rate limit and a default model in settings.
- 3
Integrate and monitor
Call the endpoint from your ITSM workflows and watch usage daily.
Control
Built so you stay in charge
Keys you control
API keys are shown once at creation and stored only as a hash. Revoke any key instantly from Settings.
Spend you can see
Every request carries its own receipt: tokens in, tokens out, latency and price. Nothing is billed that you cannot trace.
Limits you set
Per-account rate limits and a free monthly allowance keep experiments safe before they become production.
Data you decide on
You choose exactly what you send. Repeat prompts can be served from your cache, and cache hits are never billed.
For developers
One endpoint. Every model.
Send a prompt and a capability, or name a model. One bearer key, one small JSON body, and a full receipt on every response.
curl https://YOUR-DOMAIN/api/ainf/inference \
-H "Authorization: Bearer $AI_INFERENCE_KEY" \
-H "Content-Type: application/json" \
-H "Idempotency-Key: order-1042" \
-d '{"prompt":"Summarise this ticket in one line","capability":"chat"}'Questions, answered
How are API keys stored?
A key is shown once at creation. We keep only a hash, so it cannot be read back. Lose it and you create a new one.
Can I see every request my organisation makes?
Request history records each call's model, tokens, latency, price and a short preview, so you can audit usage.
Can I send internal data?
You control exactly what each prompt contains. Review your own data handling policies before sending sensitive content.
What if a model goes down?
The router retries the request on another eligible model automatically.
Bring your AI usage in from the shadows
One endpoint, hashed keys, rate limits and history. Set it up this afternoon.