Products
Every way to run AI with AxForge
Call a model, keep your coding tool, have your calls served first, rent the machine, or have us run it for you. One account and one balance, on machines we own in Europe.
All products
Pick the amount of control you want
-
Managed Endpoint
The same API and models, with your calls run ahead of shared traffic, at a fixed price.
from €209.30 / month
-
Managed GPU
Tell us what to run. We deploy and operate it on a dedicated GB10, as your own endpoint.
€518.30/month for the dedicated GB10, the service quoted per deployment
-
Bare metal
A whole GB10 machine with root, on its own network, by the month.
from €552.50 / month
-
Xev decisions
Yes or no, a choice or a score, with probabilities. Jev-compatible.
€0.022 / 1M input tokens, output free
-
OCR, every EU language
PDFs, scans and photos to text, billed per page.
€0.003 / page
-
AI agents for finance
Agents that do the work for regulated finance teams, built and run with you.
Scoped with you
-
Serverless Models
Open models behind one OpenAI-compatible API: chat, embeddings, images, speech and music.
From €0.12 / 1M input tokens
-
AxForge Coder
Our models in the coding tool you already use, with tokens every month.
From €6/month
-
GPU rental (dedicated GPU machine)
A dedicated DGX Spark (GB10) with full SSH, by the hour, week, month or year.
DGX Spark (GB10) from €0.69/hour
Not sure which?
A few questions decide it
Do you want tokens or a machine? Tokens: Serverless Models, or Coder for a coding agent. A machine: GPU rental or bare metal.
Who runs it? You, with full SSH access, or AxForge: Managed GPU deploys and operates your model, Managed Endpoint serves your calls first on ours.
Where must the data stay? In the EU, whatever you pick: where data lives, what is kept and the DPA.
Still unsure? Talk to an engineer, or ask the community.