Slash AI token costs by 70%
Join us at our hospitality suite, minutes from STATION-Berlin for 1:1 meetings with our Solution Architects or Executives for either live demos or specific conversations on your AI token spend and how you can leverage local-first inference to slash your token costs.
Spectro Cloud at NVIDIA GTC Berlin 2026
4 minutes from STATION-Berlin
Local inference takes over Gibson Bar
Just minutes from STATION-Berlin on Stresemannstraße, join us at our hospitality suite for live, personalized demos.
10:00 to 18:00
08:00 to 17:00
Join the team behind Spectro Cloud PaletteAI Inference Launchpad for 1:1 discussions or demos. Our hospitality suite offers an opportunity for targeted discussions to determine how you can leverage open weight models and owned hardware and slash token costs.
Want to see how the product works? Our demo stations are equipped to do exactly that.
Connect with us over Canapés.
Wanna DJ? While you're there, join our live DJ to learn how to mix cool beats!
Drop-ins are welcome but please book ahead to secure a meeting slot.
.png)
Read up
Certified for NVIDIA Run:ai, and a launch partner for the open-source release
Run:ai handles the part most platform teams find hardest: sharing expensive GPUs fairly across teams that all want them at once. Palette now runs Run:ai as a certified configuration, so your data scientists get their quotas and your platform team keeps control of the cluster underneath. We're also a launch partner for the open-source release.
PaletteAI is NVIDIA AI Cloud-Ready validated
PaletteAI has been validated against NVIDIA's AI infrastructure reference architecture under the AI Cloud-Ready ISV program. If you are building on NVIDIA Cloud Partner infrastructure, that means the Kubernetes and platform layer is a known, tested quantity rather than something your team integrates and validates on its own time.
%20(1).png)
Off the Grid "After Party" with WEKA
Join us and WEKA for an Off the Grid "After Party" aboard the Van Loon with live entertainment and refreshments.