Mar 10, 2026
🚀 Power Up: 2x Memory & New 32GB XLarge
To support your most ambitious AI projects, we’re delivering a significant performance boost. Enjoy doubled memory across standard tiers and meet our new powerhouse—the 32GB XLarge tier.
⚡ 2x Memory Boost, Same Price We’ve optimized our container specifications to a 1:4 (vCPU to Memory) ratio. This update doubles your memory capacity for a more stable experience—at no extra cost. (Note: Micro and GPU tiers remain unchanged in this update.)
Small
2 vCPU / 2GB
1 vCPU / 4GB Memory
6.80
Medium
4 vCPU / 4GB
2 vCPU / 8GB Memory
13.60
Large
8 vCPU / 8GB
4 vCPU / 16GB Memory
27.20
XLarge 🌟
-
8 vCPU / 32GB Memory
54.40
How to apply: New containers get this automatically. For existing containers, simply ‘Restart’ to apply the new specs. Your files and configurations will remain safe.
✨ New Performance Tier: XLarge (8 vCPU / 32GB) Exclusively for membership users, the XLarge tier is purpose-built for heavy-duty AI workloads that demand peak performance:
LLM Training: Efficiently fine-tune and run large language models.
Complex Pipelines: Execute multi-model AI workflows without memory bottlenecks.
Massive Data: Handle high-memory datasets with 32GB of dedicated RAM.
Minor Changes
New Features
gemini-3.1-pro Support: Added gemini-3.1-pro to the list of available models in the Side Chat (Workspace), providing more refined and up-to-date intelligence for your tasks.
Changes
Email Compliance: Added company legal names and addresses to the footer of all Arkain system emails for enhanced transparency and regulatory compliance.
Bug Fixes
Container Settings Page: Fixed control buttons overlapping the Side Chat (Dashboard) input box on mobile.
Deprecated
gemini-3.0-pro: Removed gemini-3.0-pro from the Side Chat (Workspace) model options, officially replaced by the improved 3.1 version.
Last updated