DigitalOcean officially launched its AI-Native Cloud at the Deploy 2026 conference, as detailed in a May 5, 2026 blog post by Chief Product & Technology Officer Vinay Kumar. The platform is structured around five integrated layers—Infrastructure (with owned silicon), Core Cloud, Inference Engine, Data & Learning, and M The post emphasizes that the stack is "independently useful and beautifully integrated," combining GPU-level economics with a managed agent runtime, vector and knowledge base services, and a kernel-optimized inference endpoint supporting open and frontier models. The author frames it as an opinionated, open-at-every-la

DigitalOcean's Inference documentation, last verified August 24, 2026, defines a five-tier quota system per account controlling Agents, Knowledge Bases, Requests Per Minute, and Tokens Per Minute. Tier 1 caps agents/Knowledge Bases at 15 with 120 RPM and 500K-750K TPM, while Tier 5 raises those limits to 120 agents/Kno The docs also clarify that all tiers except Tier 1 and Tier 2 can access every model in the catalog, while the lower tiers are blocked from Anthropic models and most OpenAI models, retaining only gpt-oss-120b and gpt-oss-20b. Bring Your Own Models imports are capped at 10, 50, or 100 models by tier, accept only Safeten

DigitalOcean has made GLM-5.3-Flash from Z.ai generally available on its Inference Engine, replacing the earlier Ox Alpha preview. The model is a 320-billion-parameter mixture-of-experts that activates 18B parameters per token and uses a hybrid sparse-and-linear attention architecture with IndexPool key-vector compress DigitalOcean cites benchmark gains of 84.3 on Terminal-Bench 2.1 and 63.4 on DeepSWE v1.1 (up from 46.2 on GLM-5.2), alongside improved agentic and coding capabilities. GLM-5.3-Flash is exposed via Serverless Inference, can be added to the DigitalOcean Inference Router routing mix, evaluated through the Model Evaluatio

DigitalOcean's Q2 2026 earnings call transcript (dated August 4, 2026) reports revenue of $281 million, up 29% year over year and exceeding the high end of guidance, with annual run-rate revenue reaching $1.13 billion (+29% YoY) and incremental ARR of $93 million, the highest quarterly addition in company history. AI C The call cites over 6,000 Inference Engine customers since launch, token volume increasing 30-fold over the prior 60 days, and open-weight model usage rising to 75% of total token volume from approximately 15% at initial launch. Large customer cohorts expanded sharply, with $1M+ ARR customers growing 214% YoY (23% of t

The latest company information, including net asset values, performance, holding & sectors weighting, changes in voting rights, and directors and dealings., The latest company information, including net asset values, performance, holding & sectors weighting, changes in voting rights, and directors and dealings.

BROOMFIELD, Colo., April 28, 2026--DigitalOcean (NYSE: DOCN) today announced the launch of its Inference Engine, a set of new production capabilities that give AI builders exceptional performance and unified control over how they run, scale, and optimize inference workloads. The announcement comes ahead of DigitalOcean

DigitalOcean has opened a private preview of Managed Agents Runtime Services (M.A.R.S.), a fully managed environment for running AI agents at scale. M.A.R.S. combines Harness Runtime, which provides managed agent execution in isolated Firecracker microVMs with durable sessions, and Action Gateway, which governs tool, A Rather than forcing a proprietary framework, M.A.R.S. accepts environment templates that package the developer's chosen harness, dependencies, tools, and configuration. Out of the box it supports Claude Code, Codex CLI, and OpenCode, as well as agents built with LangGraph or CrewAI, giving teams a managed path to run l

Combining engineering expertise and cloud innovation to accelerate AI adoption BROOMFIELD, Colo.–(BUSINESS WIRE)–DigitalOcean (NYSE: DOCN), the comprehensive agentic