DigitalOcean

Claude Fable 5.1 is now available through the DigitalOcean Inference Engine, day zero. ⚡️ https://do.co/4yk6I7M It's Anthropic's most advanced model for coding and knowledge work, now with 75% cheaper cache-read pricing, making long-running agents that replay large context far more economical. Available via Serverless Inference, Inference Router, model synthesis, and Evaluations: fully managed, usage-based pricing, no separate Anthropic contract.

Your coding agent dies the moment you close your laptop. So we built DigitalOcean Managed Agents Runtime Services (M.A.R.S.). One doctl command drops Claude Code, Codex, OpenCode, or your own LangGraph agent into an isolated sandbox. Pause it, close everything, resume tomorrow. Workspace and full conversation intact. Checkpoints snapshot the machine and agent memory together. Fork one session into three, try three approaches on the same refactor, keep the winner. Action Gateway handles tools: GitHub, Jira, Stripe, 1,000+ more behind a single MCP endpoint. Your keys stay in Secrets Manager until a tool runs, never touching the model. Sign up for the private preview today. 🔗 https://do.co/4cokbmt

Our Solutions Architect Darian Wilkin put Z.ai's GLM-5.3 to the test in Flotilla, a naval resource-and-diplomacy simulation where four LLMs write and rewrite their own ship-control code across a five-game series, this time against Kimi, DeepSeek, and Qwen. A few standout moments from GLM-5.3's run: • Won the series 3-2, closing game five with its highest score of the series • The only model to use dispatch cutters, relaying new orders to ships already out at sea • Diagnosed and rewrote its own broken ship code mid-run rather than stalling • Negotiated, broke, and re-negotiated non-aggression pacts as the game state shifted, timing an offensive push to when it had the advantage Try it yourself: https://lnkd.in/gQgqkCQj

Curious about GLM-5.3-Flash? Our Senior Developer Advocate, Ryan O'Connor, walks through what makes this model from Z.ai worth a look for agentic workloads. A few highlights: • Scores 57 on the intelligence index, delivering top-tier agentic coding performance at a fraction of typical cost • First natively multimodal model in the GLM series, enabling browser and computer use so it can inspect UIs and critique code visually • 1M token context window, MoE architecture (320B total / 18B active parameters per token) • Nearly doubles GLM-5.2's performance on AutomationBench It runs fully managed on DigitalOcean's Inference Engine; swap one endpoint and start testing it against your own workloads.

This summer we patched two hypervisor-class vulnerabilities across the fleet. Read about it in a new blog post from DigitalOcean's Deputy CISO, Timothy Lisko: https://do.co/4iA3qbx The first was Januscape (CVE-2026-53359), which let a malicious guest escape to the host. That is the most serious kind of vulnerability a cloud provider can have. The team built and validated the first livepatch 45 minutes after answering the page, protected thousands of hypervisors within 48 hours, and finished the fleet in 8 days. A few hundred hosts were running an older kernel that couldn't take the livepatch, so the team evacuated and rebuilt those instead. The second affected the Safe RET speculative-execution mitigation on AMD hardware, where interrupt timing could weaken the protection. It couldn't be livepatched because of how the mitigation is compiled, so every affected host needed a kernel update and a reboot. That came to ~1,600 hypervisors in 12 regions, two-thirds cleared in the first three days and the rest finished within the coordinated disclosure window. Neither response caused confirmed customer-facing impact.

GLM-5.3 from Z.ai is now on DigitalOcean Inference Engine, day 0. https://do.co/4wSFx2o It's Z.ai's strongest agentic coding model yet, and the most efficient. At max effort it hits 34.5% on Z.ai's Code Bench using ~75K output tokens per task, up from 23.4% at 96K for GLM-5.2, so each completed task costs less even where per-token rates match. The 1M-token context window holds an entire repository in one request. Run it on Serverless Inference, route to it alongside 75+ models with Inference Router, combine it with other models through model synthesis, or test it against your current model with Evaluations before switching production traffic. Fully managed, one bill, no separate Z.ai account.

Z.ai GLM-5.3-Flash, the artist formerly known as Ox Alpha, is now available through DigitalOcean Inference Engine. 🆕 https://do.co/45O4zoz The first natively multimodal model in the GLM-5 series, GLM-5.3-Flash can read images and video alongside text, with a 1M-token context window and hybrid attention architecture that reduces compute and KV cache size by 3-4x. Put GLM-5.3-Flash to work: 🧠 Add GLM to your routing mix for more intelligence per dollar. 🔀 Synthesize across GLM and other models for better quality. 📊 Evaluate GLM and other models to find the best fit for your workloads. ⚡ Run it through Serverless Inference with fully managed infrastructure and usage-based pricing.

Ever run the same LLM inference benchmark twice and get two different "best" numbers? So did we: 12 minutes apart, same system, no changes, and p99 latency still moved 26%. 🤔 Turns out most benchmarks tilt on 15 setup choices most people never mention. Our new guide breaks down the levers, shows what they did to a live run on DigitalOcean Serverless Inference, and gives you a checklist to pressure-test any number before you trust it, yours or someone else's. 🔗 https://do.co/4qBXWiC