Serverless Inference now available on GenAI Platform
With serverless inference, you can now integrate AI models from OpenAI, Anthropic, Meta, and more through a single API, no infrastructure provisioning required. It offers reliable endpoints, automatic scaling, and centralized billing, making it easy and cost-effective to embed AI into your applications. To learn more check out our announcement blog, or get started with this tutorial →
New observability, monitoring, and database hardening updates for Managed Databases
Observability and monitoring are crucial for visibility into your tech stack. That’s why we’re releasing three new advancements in this space for our Managed Databases platform. Learn more →
New Spaces features make it easier to stay secure, compliant, and in control
Managing object storage at scale can be hard, from tracking file access through CDNs to setting up tools like the AWS CLI and keeping everything secure and compliant.
- Spaces access logs (now generally available) provide detailed, timestamped records of all file activity—including CDN traffic—to help you monitor usage, spot unusual patterns, and support compliance needs.
- The connection wizard simplifies setup for third-party tools, preventing misconfigurations and rate-limiting errors.
No extra cost, just better visibility, less risk, and smoother workflows. Whether you’re a developer, DevOps pro, or part of a growing team, these updates keep you in control. Learn more →
New networking features to strengthen your cloud deployments
We’ve rolled out three powerful updates to enhance the resilience and flexibility of your cloud infrastructure and to improve network control. Check them out:
- Partner Network Connect now supports high availability and DOKS: Help ensure uninterrupted connectivity between multi-cloud environments with automatic failover. If one link goes down, traffic seamlessly switches to the other without customer impact, an ideal feature for critical workloads like streaming, gaming, and ML pipelines. Securely connect VPC-native DOKS clusters with Kubernetes workloads across multi-clouds for low-latency, hassle-free communication.
- Reserved IPv6 on Droplets: Say goodbye to IP address changes! With this update, you can reserve and retain your IPv6 addresses across Droplet reboots or replacements, making your infrastructure more stable and predictable, particularly for apps with hardcoded IPs.
- Bring Your Own IP (BYOIP): Preview this feature to migrate to DigitalOcean without disrupting your IP reputation or forcing costly reconfigurations. Import your own public IPv4 blocks and assign them to Droplets or Kubernetes nodes, avoiding downtime and simplifying the migration process.
Learn more →
New GPU Droplets powered by AMD
AMD Instinct™ MI300X delivers leadership performance for the newly exploding demands of generative AI. Use-cases for these GPUs include large model training, fine-tuning, inference, and HPC.
Key benefits:
- Exceptional memory performance and bandwidth: 192 GB of HBM3 memory with a peak theoretical bandwidth of 5.3 TB/s.
- Optimized for Generative AI and HPC: Built on the CDNA 3 architecture, which is designed for GenAI and HPC.
- Value: Available at a competitive price point ($1.99 GPU/hr/on-demand) for a HPC GPU.
Learn more →
