Products · Managed Endpoint

Managed Endpoint — your calls served first

The same OpenAI-compatible API and models as Serverless, with your chat calls run ahead of shared traffic. Choose how many run first at once — 2, 4 or 6 — and pay a fixed price by card, monthly, every 6 months or yearly.

Available eu-se-1 · Stockholm api.axforge.ai/v1

Choose a plan Compare with Serverless from €299 / month, no VAT charged — tokens at the serverless price, from your balance.

What you get

First in line, nothing else to change

Priority	Chat calls from your Managed keys run before shared API traffic — as many at once as your plan.
API	The same address and models as Serverless — api.axforge.ai/v1. You choose which keys are Managed in the console; your code changes only its key.
Plans	Managed 2, 4 or 6 — how many calls run first at once. A fixed price for the period, by card.
Tokens	Usage at the serverless price, from your balance — from €0.32 / 1M input tokens. All prices.
Billing	Monthly, every 6 months or yearly — renews by itself, cancel any time. A larger plan works the moment you choose it and your next invoice bills it; a smaller one from your next invoice. When a plan ends, switch its keys back to Shared — a Managed key without a plan is refused.
Region	Stockholm, Sweden (eu-se-1) — inference runs in-region.
Data	Zero prompt retention — prompts and completions are processed in memory and never stored. Trust Centre.

Plans

Managed 2, 4 or 6

Plan	Monthly	6 months Save 15%	Yearly Save 30%
Managed 2 2 calls first at once	€299	€1,524.90€254.15 / month	€2,511.60€209.30 / month
Managed 4 4 calls first at once	€499	€2,544.90€424.15 / month	€4,191.60€349.30 / month
Managed 6 6 calls first at once	€699	€3,564.90€594.15 / month	€5,871.60€489.30 / month

Each price is the whole period, no VAT charged; tokens are billed as on Serverless.

How it works

Three steps

1 · Choose a plan	In the console — monthly, every 6 months or yearly, paid by card.
2 · Make a key Managed	On API keys, switch a key to Managed. Its chat calls run first from then on.
3 · Call the same address	Your code keeps its address and models — only the key is the Managed one.

First request

The same call, served first

# a key set to Managed, from API Keys
export AXFORGE_API_KEY="YOUR_KEY"

curl https://api.axforge.ai/v1/chat/completions \
  -H "Authorization: Bearer $AXFORGE_API_KEY" \
  -H "Content-Type: application/json" \
  -d '{"model":"qwen3.8-27b-nvfp4","messages":[{"role":"user","content":"Hello"}]}'

# a key set to Managed, from API Keys
$env:AXFORGE_API_KEY = "YOUR_KEY"

$call = @{
  Method      = "Post"
  Uri         = "https://api.axforge.ai/v1/chat/completions"
  Headers     = @{ Authorization = "Bearer $env:AXFORGE_API_KEY" }
  ContentType = "application/json"
  Body        = '{"model":"qwen3.8-27b-nvfp4","messages":[{"role":"user","content":"Hello"}]}'
}
(Invoke-RestMethod @call).choices[0].message.content

REM a key set to Managed, from API Keys
set AXFORGE_API_KEY=YOUR_KEY

curl.exe https://api.axforge.ai/v1/chat/completions ^
  -H "Authorization: Bearer %AXFORGE_API_KEY%" ^
  -H "Content-Type: application/json" ^
  -d "{\"model\":\"qwen3.8-27b-nvfp4\",\"messages\":[{\"role\":\"user\",\"content\":\"Hello\"}]}"

Your first call, step by step — every language and shell

View quickstart · Connect your tools

Which product

Serverless, Managed Endpoint or Managed GPU

Serverless	The shared API — pay per token, nothing to set up.
Managed Endpoint	The same API with your calls first — a fixed price by plan, tokens as on Serverless.
Managed GPU	A whole machine for your traffic only — AxForge operates it, quoted per deployment.

Keep reading

Serverless Models Model catalogue Pricing Quickstart Managed GPU Trust Centre