Date: March 17, 2026
We are thrilled to announce the official completion and launch of the new Gnoppix API Server. To ensure a premium experience for our global community, we have deployed additional nodes worldwide, providing ultra-low latency regardless of your location. Thanks to nova provide the hardware.
You do not need anymore country specific settings. The cluster just routes you to the shortest distance. the endpoint settings are : https://api.gnoppix.org
Our Commitment to Open Source
As industry giants like Meta shift their strategies toward closed-source models, Gnoppix remains committed to true digital sovereignty. We are doubling down on our support for high-performance, open-source alternatives.
We are excited to highlight the inclusion of Qwen models, which have become a community favorite for their transparency and incredible reasoning capabilities.
New Models for Gnoppix Members
We have curated a diverse selection of powerful models to suit every workflow. The following models are now available for our members:
| Provider | Model Name |
|---|---|
| NVIDIA | nemotron-3-nano-30b-a3b:free |
| NVIDIA | nemotron-3-super-120b-a12b:free |
| NVIDIA | nemotron-nano-9b-v2:free |
| Qwen | qwen3-next-80b-a3b-instruct |
| Qwen | qwen/qwen3-next-80b-a3b-instruct:free |
| Qwen | qwen/qwen3-coder:free |
google/gemma-3n-e2b-it:free |
|
| Mistral | mistralai/mistral-small-3.1-24b-instruct:free |
| Meta | meta-llama/llama-3.3-70b-instruct:free |
| Meta | meta-llama/llama-3.2-3b-instruct:free |
| Nous | nousresearch/hermes-3-llama-3.1-405b:free |
Note: Access to these “free” tier models requires an active Gnoppix Membership subscription.
Important: Action Required for Existing Members
To integrate with the new infrastructure, all existing members must generate a new API key. This can be done quickly via our Self-Service Portal.
Get started today: https://landing.gnoppix.org
Thank you for supporting uncensored, open-source AI.
Since EU haven’t changed their AI ACT we still cannot support Users located within the EU.
P.S. the same limitation applied we share 19 queries / second and 1000 prompt per day.