​
Model APIs Updates

​
New Features

  • Novita LLMs Dedicated Endpoints Fully Managed Deployment for Your Custom AI Models High cost performance
    • Deploy available models and LoRA adapters on customizable GPU endpoints with per-hour billing.
    • Lower prices, No hard rate limits, Cheaper under high utilization, Custom base models and multiple LoRAs, Predictable performance unaffected. Learn more about Pricing and read this Blog to learn how to deploy a custom base model with Novita LLMs Dedicated Endpoints.
    Updated time: July 4, 2025

​
Agent SandBox Updates

​
New Features

  • Agent Sandbox Is Now Live Agent Sandbox is a high-performance cloud runtime purpose-built for AI Agent workloads, designed to support the full lifecycle of autonomous agents. Key capabilities include:
    • Millisecond-level startup, optimized for high-concurrency tasks
    • Multi-language support, including Python, JavaScript, and C++
    • Enables code execution, web access, and system-level interactions
    • Per-second billing for vCPU and memory — flexible, cost-efficient usage
    As the infrastructure foundation for the runtime era of AI Agents, Agent Sandbox empowers the next generation of generative agents with fast, scalable, and reliable execution capabilities. Visit the Documentation Center to learn more. Updated time: June 30, 2025

​
Platform Service Updates

​
Improvements

  • In-App Messaging Now Live for Critical Notifications Our new in-app messaging system is now available, allowing you to receive important platform-level updates in one place, including:
    • GPU failures, service interruptions, and operational alerts
    • Account-related notices, such as coupon delivery or login activity
    • Promotional events and platform campaigns
    All messages can be viewed in your Message Center in the console. Stay informed, effortlessly. Updated time: July 1, 2025