Ce que DigitalOcean a publié en juin 2026

Built for Mass Scale: Hard-Won Lessons from Teams Running High Volume Inference Workloads in Production

DigitalOcean hosted a panel at Deploy 2026 featuring Workato, Hippocratic AI, and ISMG, highlighting challenges in scaling AI inference workloads to production. Key issues included P99 latency spikes,…

DigitalOcean launches MCP tools for AI agent and model evaluations

DigitalOcean introduced a suite of MCP tools for its GenAI platform, enabling end-to-end evaluation of AI agents and raw models. The tools allow users to manage datasets, create test cases, run evalua…

Run Codex in the cloud – DigitalOcean for Codex is now available

DigitalOcean introduced a Public Preview integration with Codex, enabling developers to provision persistent cloud development environments (Droplets) directly from Codex using natural language. The p…

Making sure you're not a bot!

DigitalOcean to charge 18% VAT in Peru starting June 2026

DigitalOcean will begin charging 18% VAT to customers with a tax location in Peru effective 1 June 2026, with charges appearing on invoices from 1 July 2026. Customers with a valid Peruvian RUC can av…

Server-Side Tools Are Now Available for DigitalOcean Inference Engine

DigitalOcean introduced Server-Side Tools for its Inference Engine, enabling AI models to access web search, web fetch, Knowledge Bases, MCP servers, and Anthropic/OpenAI tools directly within inferen…

DigitalOcean to force MySQL 8.0 upgrades to 8.4 by Oct 30 2026

DigitalOcean announced MySQL 8.0 will reach end of life on October 30, 2026, triggering forced upgrades to MySQL 8.4 for all managed clusters during maintenance windows. No downtime is expected, but l…

Now Available: Anthropic Claude Fable 5

nvidia/NVIDIA-Nemotron-3-Super-120B-A12B-NVFP4like346FollowNVIDIA59.8k

NVIDIA released Nemotron-3-Super-120B-A12B-NVFP4, a 120B-parameter LLM optimized for agentic reasoning and high-volume workloads like IT automation. It uses a hybrid LatentMoE architecture with Mamba-…

The Inference Alpha: Maximizing Frontier Models on AMD

DigitalOcean demonstrated order-of-magnitude performance improvements for frontier LLMs on AMD GPUs through deep software stack optimizations, including MXFP4 quantization, MLA attention, kernel fusio…

NVIDIA DGX B300

NVIDIA introduced the DGX B300, a new AI supercomputer powered by Blackwell Ultra GPUs, delivering 1.5x denser FP4 performance and 2x attention performance over DGX B200. The system supports real-time…

Kimi K2.6: From Code to Creation, From One to Many

Moonshot AI released Kimi K2.6, an open-source AI model featuring state-of-the-art coding, long-horizon execution, and agent swarm capabilities. The model enables single-prompt creation of full-stack …

Trinity Models

Arcee AI introduced Trinity, a family of open-weight language models designed for multi-turn conversations, tool use, and structured outputs. The lineup includes Nano (6B/1B active), Mini (26B/3B acti…

Janet Harrah

What We Learned Hiring 33 Engineers in Two Weeks

DigitalOcean rapidly hired 33 engineers in Seattle by redesigning its interview process around AI-native workflows, replacing traditional whiteboard tests with a three-hour build session using AI tool…

Claude Fable 5 and Claude Mythos 5

Anthropic released Claude Fable 5, a state-of-the-art AI model excelling in software engineering, vision, and scientific research, alongside Claude Mythos 5, a variant with lifted safeguards for cyber…

Model Evaluations: Prove Your Routing Policy Actually Works

DigitalOcean introduced Model Evaluations in Public Preview, enabling users to test and compare inference models, routing strategies, cost, latency, and output quality before deploying to production. …

Now Available: NVIDIA Nemotron 3 Ultra

The Team Behind Deploy: Shipping AI, the DigitalOcean Way

DigitalOcean unveiled its AI-Native Cloud, a five-layer stack for AI-native companies, featuring over 15 product launches including Inference Router, Dedicated Inference, Managed Weaviate, Knowledge B…

Powering the Inference Era: Inside the DigitalOcean Data & Learning Layer

DigitalOcean introduced a unified Data & Learning layer to unify structured and vector data for AI-native applications. Key updates include PostgreSQL and MySQL Advanced Edition (Public Preview) for h…

Now Available: Claude Opus 4.8

Open by Design: How NVIDIA and DigitalOcean Are Building the Stack for the Always-On Agentic Era

DigitalOcean and NVIDIA highlighted the importance of open-source AI models and infrastructure for building agentic systems in a Deploy 2026 session. They emphasized model flexibility, evaluation stan…

DigitalOcean Serverless Inference: A Deep Dive

DigitalOcean introduced Serverless Inference, a fully managed API-first platform supporting 30+ foundation models across text, code, vision, image, video, and speech. It offers single-endpoint access,…

The Inference Tax: How Prefix-Aware Routing Eliminates the Hidden Cost of LLMs at Scale

DigitalOcean unveiled a prefix-aware routing and caching mechanism to eliminate redundant LLM inference costs, targeting the 'prefill tax' where identical system prompts and shared contexts are recomp…