Call a model, not a cluster
One OpenAI-compatible endpoint for leading open models, billed by the token. No GPU to reserve, no capacity to forecast — ideal for product teams shipping chat, RAG and code features.
Pay per token, no minimum
Input and output tokens are priced separately. Prepaid balance or postpaid invoicing, with usage you can audit line by line.
Developer
Ideal for feature validation and small-scale integration, with free credit for the first month.
- Pay per token, no minimum
- Shared inference queue
- Free first-month credit
- Community support
Standard
Built for production, with a 99.9% uptime SLA and a priority inference queue.
- 99.9% uptime SLA
- Priority inference queue
- Rate limits and monthly quotas
- Email support, 24h response
Enterprise
Dedicated inference capacity with a private gateway, supporting industry-model customisation and on-premise rollout.
- Dedicated GPU capacity
- Private token gateway in your VPC
- Custom industry model tuning
- Named architect, 24/7 support
Token pricing in USD per 1,000 tokens. Enterprise tiers are quoted on committed monthly volume.
Three lines to your first completion
Point the OpenAI SDK at our base URL and keep your existing code.
curl https://api.warnertelecom.co.nz/v1/chat/completions \
-H "Authorization: Bearer $WT_API_KEY" \
-H "Content-Type: application/json" \
-d '{
"model": "qwen2.5-72b",
"messages": [
{"role": "user", "content": "Summarise this support ticket."}
],
"stream": true
}'A gateway you can put your name on
Pay per token, no minimum
Input and output priced separately, prepaid or postpaid, with a usage log you can reconcile against your own metrics.
Unified model access
One key for Qwen, DeepSeek and the Llama family plus our own tuned industry models — switch by scenario, not by vendor.
Private token gateway
Enterprise customers can run a dedicated inference gateway so token usage and call logs stay inside their own VPC for audit.
Rate limits and quotas
Set call rates and monthly token quotas per application or team, so one runaway process can't blow the budget.
Ready to test it?
Tell us the workload and we'll issue a sandbox key with credit for your first month.
Request sandbox access