Cloud-first · small teams
- 8-core CPU with integrated 50 TOPS NPU
- 32 GB unified memory
- 1 TB NVMe storage
- ~28 tok/s on Qwen 2.5 7B
- Near-silent (fanless)
- 3-year warranty
Three appliances. Monthly LLM allowance included, or bring your own keys. All prices inc VAT.
Cloud-first · small teams
Hybrid · mid-size offices
Regulated industries · air-gap available
Prices subject to final supplier and margin sign-off · UK pounds sterling
The appliance, PSU, mounting kit, and cables — shipped pre-configured.
Firmware, redaction pipeline, audit log, chat UI, and browser extension.
Hardware warranty with next-business-day replacement in the UK.
For the lifetime of the appliance — via your tunnel or USB on air-gapped sites.
60 minutes with your IT team: install, SSO, and your first custom rule pack.
Real people in the UK by email and phone. No outsourced tier-one.
Use the included allowance through Trinito-managed cloud, or bring your own OpenAI, Anthropic, or Google keys at no extra charge.
| Plan | Included monthly tokens | Overage rate | BYO-key option |
|---|---|---|---|
| Trinito Compact | 5 million | £0.60 per 1M tokens | Included |
| Trinito Standard | 15 million | £0.50 per 1M tokens | Included |
| Trinito Sovereign | Custom | POA | Included |
Tokens are pooled across your organisation. Provider mix and allowance pricing are reviewed quarterly.
Many customers buy both. The Gateway protects outbound prompts to public LLMs; rental servers run your own LLMs on-prem.
Tell us headcount, which tier looks right, and any compliance constraints. We will confirm price, lead time, and a demo slot.