Dedicated model · deployable on request

Qwen3.6 35B A3B — dedicated hosting in Europe

A mixture-of-experts Qwen with roughly 3B active parameters per token, validated on AxForge hardware and deployed on a dedicated DGX Spark for your traffic only. Not on the shared API today.

On request eu-se-1 · Stockholm 30.7 tokens/s measured
Talk to an engineer Qwen3.8 27B — served now Monthly rental. Pricing on request.

Specifications

The model

ModelQwen3.6 35B A3B — mixture-of-experts, latest Qwen generation
Served model nameqwen3.6-35b-a3b-nvfp4
ArchitectureMoE — 35B total, roughly 3B active parameters per token
Context window65,536 tokens
QuantizationNVFP4
AvailabilityDeployable on request — not on the shared API today
HardwareNVIDIA DGX Spark (GB10, 128 GB unified memory) — owned and operated by AxForge
Regioneu-se-1 · Stockholm, Sweden
PricingOn request — talk to an engineer

Performance

Measured, not quoted

MetricValueCondition
Decode speed30.7 tokens/ssingle stream, MoE with ~3B active parameters

Measured on our production DGX Spark node, single-stream, 2026-08. Faster single-stream than the dense Qwen3.8 27B we serve (15.9 tokens/s measured) because only ~3B parameters are active per token. For quality benchmarks, see the official model card — we publish only numbers we measured ourselves.

Deployment

What "deployable on request" means

SystemA dedicated NVIDIA DGX Spark (GB10, 128 GB unified memory), reserved for you
Commercial modelMonthly rental
EndpointOpenAI-compatible /v1 on your own machine — only your traffic
LocationHosted in the EU (eu-se-1 · Stockholm, Sweden)
Data handlingZero prompt retention, same policy as the shared API

The model is validated on AxForge hardware and brought hot on your system. After deployment — and only then — your endpoint speaks the OpenAI API:

Post-deployment sketch — not the shared API
# After your dedicated node is live. This does not work on api.axforge.ai today.
from openai import OpenAI

client = OpenAI(base_url="https://<your-endpoint>/v1", api_key="<your-key>")
r = client.chat.completions.create(
    model="qwen3.6-35b-a3b-nvfp4",
    messages=[{"role": "user", "content": "Hello from your own machine"}],
)

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. The full policy is at axforge.ai/privacy.

FAQ

Qwen3.6 35B A3B hosting — common questions

Is Qwen3.6 35B A3B on the shared AxForge API?

No. It is not on the shared API today. It is deployable on request: validated on AxForge hardware and brought hot on a dedicated DGX Spark for your traffic only. The shared API serves Qwen3.8 27B.

What does A3B mean?

Qwen3.6 35B A3B is a mixture-of-experts model: 35B total parameters, roughly 3B active per token. That is why its single-stream decode is faster than a dense model of similar size.

How fast is Qwen3.6 35B A3B on your hardware?

We measured 30.7 tokens/s single-stream on our production DGX Spark node, 2026-08. We publish measured numbers only.

What does a dedicated Qwen deployment look like?

A dedicated NVIDIA DGX Spark (GB10, 128 GB unified memory) rented monthly, running Qwen3.6 35B A3B behind an OpenAI-compatible endpoint on your own machine, hosted in the EU with zero prompt retention.

What does dedicated Qwen hosting in the EU cost?

Pricing on request — talk to an engineer and we will quote the monthly rental.

Which Qwen model can I use through the API today?

Qwen3.8 27B is served now on the shared API at €0.29 per 1M input tokens and €1.77 per 1M output tokens (launch pricing).

Ready to build on EU inference?

Request deployment Talk to an engineer

Explore

All models

© 2026 AxForge · EU-hosted AI infrastructure Pricing Docs Trust Privacy Terms