Platform · Available

EU AI API: OpenAI-compatible, hosted in Europe

One API for chat, embeddings, images and audio: OpenAI-compatible, served from EU regions on hardware AxForge owns, with zero prompt retention.

eu-se-1 · Stockholm api.axforge.ai/v1
Get API access Talk to an engineer Create an account. Your key is in the console. 3M free tokens every 30 days with every new account.

Catalogue

Models on the serverless API today

ModelTypeContextPricing
Qwen3.8 27BLLM: chat, tools, streaming, usage262,144€0.32 / 1M in · €2.00 / 1M out (launch pricing)
Qwen3 EmbeddingEmbeddings: 1024-dim vectors32,768€0.02 / 1M
ERNIE Image TurboImage generation: dense in-image text—€0.05 / image
FLUX.2 Klein 4BImage editing: instruction-based—€0.06 / edit
MiniMax Music 3Music generation: full songs to 5 min—€0.04 / 10 seconds
WhisperSpeech-to-text: transcription endpoint—€0.005 / minute
PiperText-to-speech: fast CPU TTS—€2.95 / 1M characters

Further models (Qwen3.6 35B A3B, Qwen3 30B A3B, Gemma-4 26B, Mistral Small 3.2, SDXL) are validated on AxForge hardware and available as managed deployments on dedicated systems. See the model catalogue.

Quickstart

OpenAI-compatible: point your SDK at us

# your API key, 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":"Summarize DORA in one paragraph."}]}'
Your first call, step by step, in every language and shell
# pip install openai · the key is read from AXFORGE_API_KEY
import os
from openai import OpenAI

client = OpenAI(
    base_url="https://api.axforge.ai/v1",  # was: api.openai.com
    api_key=os.environ["AXFORGE_API_KEY"],
)
r = client.chat.completions.create(
    model="qwen3.8-27b-nvfp4",  # was: a GPT model name
    messages=[{"role": "user", "content": "Hello from Stockholm"}],
)
print(r.choices[0].message.content, r.usage)
The OpenAI SDK with AxForge: streaming, tools, images

The region is pinned on the key, so every request runs in eu-se-1 · Stockholm. The console shows the region and machine, and every response carries the model identity, so every call is verifiable.

Data & privacy

Zero prompt retention

Prompts and completions are processed in memory in Sweden, not written to disk, not logged, not retained, and never used to train anything. We keep only request metadata (token counts, timestamps, status) for billing and operations. TLS 1.3 terminates in the EU. GDPR compliant. The full policy is at axforge.ai/privacy.

FAQ

EU AI API: common questions

Is the EU AI API OpenAI-compatible?

Yes. Chat completions, streaming, tool calls and usage reporting follow the OpenAI API shape, alongside embeddings, image generation and audio (transcription and speech). The official OpenAI SDKs work by changing base_url to https://api.axforge.ai/v1.

Which models are served on the EU AI API?

Served now: Qwen3.8 27B (chat, tools, streaming), Qwen3 Embedding (1024-dim vectors), ERNIE Image Turbo (image generation), FLUX.2 Klein 4B (image editing), MiniMax Music 3 (music generation), Whisper and Piper. Further models are validated on AxForge hardware and available as managed deployments on dedicated systems.

Where is the API hosted?

Serverless inference runs in Stockholm, Sweden (eu-se-1) on NVIDIA DGX Spark (GB10) systems AxForge owns; dedicated GPU rentals run from Málaga, Spain (eu-es-1). TLS 1.3 terminates in the EU.

What does the EU AI API cost?

Qwen3.8 27B is €0.32 per million input tokens and €2.00 per million output tokens (launch pricing). Image generation is €0.05 per image and editing €0.06 per edit, embeddings €0.02 per million tokens, speech-to-text €0.005 per minute, text-to-speech €2.95 per million characters and music €0.04 per 10 seconds. The full table is above, and the price list has it all.

Are prompts retained or used for training?

No. Prompts and completions are processed in memory in Sweden, not written to disk, not logged, not retained, never used to train. Only request metadata (token counts, timestamps, status) is kept for billing and operations. See the privacy policy.

Can I pin requests to a specific EU region?

Every API key is pinned to eu-se-1 (Stockholm) and all inference for that key runs there. The console shows the region and machine your key runs on, and every response carries the model identity.

Related

Nearby on AxForge

An EU AI API you can point your SDK at today.

Get API access Talk to an engineer
© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms