LeemerLabs / Inference
One API. European compute.
Run open models through an OpenAI-compatible gateway, issue scoped keys, and inspect usage from one LeemerLabs workspace.
Gateway contractOpenAI compatible
- BASE URL
- https://api.leemerlabs.ie/v1
- AUTH
- Bearer / project-scoped key
- REGION
- European data path
- CONTROL
- Keys / scopes / usage / revoke
Access paths
Start small. Move to dedicated capacity when the work demands it.
Developer
Free
No card required
- Access to Liquid LFM2.5-350M
- 200 tok/s throughput (capable of 40K)
- 100,000 tokens / day
- 10 requests / second
- OpenAI-compatible gateway
- EU-only compute
- Community support
Team
Recommended€0.30
per 1M tokens
Pay for what you use
- Full model catalogue as it rolls out
- No daily token cap
- 100 requests / second
- Priority support (24h SLA)
- Usage analytics dashboard
- Zero-retention mode available
Enterprise
Custom
Procurement-ready
- Dedicated H200 capacity
- Custom DPA + sub-processor list
- Private model endpoints
- On-prem or VPC deployment
- 99.9% SLA with penalties
- Direct line to founder
First request
From form to first request.
01
Tell us who you are
A short form. Email, company, what you're building, rough volume.
02
Hear back same day
We reply within one working day — usually the same day — with your gateway key and onboarding notes.
03
Ship
Point your OpenAI SDK at api.leemerlabs.ie/v1. Your existing code works. Enjoy European compute.
Request a deployment
Tell us what you're building.
A founder will read this directly. Keep it short — we just need enough context to route your access key and any model-specific notes.
Want to see the gateway before you fill anything in?