DigitalOcean Release Notes

Last week we launched DigitalOcean Managed Agents in public preview: a managed runtime, governed access to more than 16,000 tools, serverless inference, and persistent memory and storage, all on one cloud.

Developers have already started thousands of agent sessions. Today we’re making the next part simpler: paying for it.

Building an agent that does something useful has gotten easy. Running it has not. A production agent needs a harness to drive the loop, a sandbox to execute code safely, inference to think, storage to remember between sessions, and tools to reach the web and other systems. Today, those come from different companies, each billing in its own unit on a separate invoice.

Here is one stack a customer described to us: OpenCode Go as the harness, Fly.io for sandboxes, AWS for storage, Fireworks for inference on open models, Anthropic for frontier models, and Parallel for web search. The harness was a flat monthly subscription. Sandbox time was billed per CPU-hour and GB-hour. Inference was per token, at a different rate for every model. Storage was per gigabyte-month plus requests. Search was per thousand queries. Six vendors, six invoices, dozens of pricing units, and a pile of glue code to hold it together.

Nobody on the team could say what a single agent run had cost.

The offerings that try to combine some of these pieces have a gap in the middle. Sandbox vendors do not run inference. Harness and model vendors do not run sandboxes or storage, and some limit agent workloads on their flat subscriptions. The hyperscalers do run everything, but they meter it as a dozen separate line items with enterprise-grade complexity to match. If you are one developer or a small team, none of these give you the thing you actually want: a price you can predict and a bill you can read.

A better way, from the people who brought you the Droplet

Fourteen years ago we put a virtual machine on the internet for $5 a month with everything included, and called it a Droplet. A generation of developers built on it because they could understand the bill. Agents are in the same place now, and we think the answer is the same. Unlike anyone else in this market, DigitalOcean prices the microVMs, storage, inference on our own hosted models, and governed access to 16,000+ tools as one simple plan with one discount. Competitors who run comparable infrastructure still sell these as separate metered line items or add-ons, for example, memory, gateway, and observability are billed separately on AWS AgentCore. DigitalOcean bundles all of it into a single plan.

Today we’re introducing Agent Droplets, one simple monthly price for DigitalOcean Managed Agents.

Pick an Agent Droplet, build your agents, and the discount applies automatically to everything they use: the compute and memory they run on, the DigitalOcean-hosted inference behind them, the storage they keep between sessions, and the tools they reach. Every agent resource is discounted, not just one layer. When your Agent Droplet allowance is used up, usage continues at list prices from your Inference and Agent Balance. Your agents keep working, and there is still only one bill.

We also include things other vendors charge for:

• No limit on the number of environments. Run one or a hundred on any Agent Droplet.
• No seat charges. An Agent Droplet covers your team’s agents, not a head count.
• No charge for enterprise security. Enterprise-grade IAM is included on every Agent Droplet.

Agents run inside dedicated microVMs that start in seconds, resume from a pause in about 300 milliseconds, and pause automatically when they are not working, so an idle agent costs almost nothing. Bring the harness you already use: Claude Code, Codex CLI, OpenCode, Hermes, or agents built with CrewAI or LangGraph.

We plan to add more resources to Agent Droplets over time.

Six vendors and six bills today, versus one Agent Droplet on DigitalOcean.

Try it free, then scale into a plan

New customers can start with a free trial: $5 in credit, no card required, enough to take an idea from zero to running this afternoon. When you’re ready for more, Agent Droplet Pro and Team apply a bigger discount to every hour, token, and gigabyte your agents use, 15% off on Pro and 20% off on Team. Moving up a tier costs more but lowers the price of everything underneath it.

[Pricing table omitted for brevity]

Example workloads and resource usage

[Example workloads table omitted for brevity]

*Figures are estimates, not an included allowance; actual usage depends on your workload. Estimates assume a usage mix of ~70% inference, ~24% compute and memory, about 4% storage and about 2% tools; inference at $0.7766 per 1M tokens (blended average for GLM 5.3); and compute on the Medium shape (2 vCPU, 4 GB).

If your Inference and Agent Balance reaches $0, running sessions pause at the next safe stopping point and their working state is preserved in a snapshot. Completed work is preserved; add funds to resume where you left off.

What $50 of agent work looks like on paper

Take the stack from earlier and give it a month of realistic work: a few agents running about two hours a day, mostly on open models, with a frontier-class model for the hard steps. Here is what arrives in the inbox at the end of the month, and what the same work looks like on Agent Droplet Pro.

[Image showing invoices comparison]

Six invoices, 17 line items, and 10 pricing units for one month of agent work, versus one line on one bill. Illustrative; competitor rates are published list prices as of September 2026, rounded.

How Agent Droplet Pro compares

Here is what the $50 a month Agent Droplet Pro gets you next to the alternatives.

[Comparison table omitted for brevity]

Competitor pricing as published in September 2026; see each vendor’s pricing page for current rates.

One balance for your agents

Agent Droplets launch alongside Inference and Agents Balance, a prepaid balance scoped only to AI products. Fund it in any amount from $5 to $500, and it’s redeemable against Harness Runtime, Action Gateway, and Serverless Inference. Standard infrastructure like Droplets, Spaces, and Managed Databases will still be billed at the end of your billing cycle as usual. Your core cloud spend keeps running exactly as it does today, completely untouched.

Fund your Inference and Agents Balance directly, and your agents draw on it automatically once your Agent Droplet allowance is used up. You never have to prepay the rest of your DigitalOcean bill to keep agents running.

Get started today

Agent Droplets are available today for DigitalOcean Managed Agents in public preview. If you already run agents on Claude Code, Codex CLI, or OpenCode, point them at DigitalOcean and they can run on an Agent Droplet with minimal setup. New DigitalOcean users get a $5 credit to launch their first agent session.

“The Droplet was never about a server or a VM,” says Vinay Kumar, Chief Product and Technology Officer at DigitalOcean. “It was about the moment you have an idea at 11 p.m. and want it running before you go to sleep. Agents should feel like that. An Agent Droplet is one subscription for everything an agent needs. We take the guesswork out of adopting new agent technologies and managing spend, so companies can start their journey with DigitalOcean and scale it with us. You pick a size and start.”

An AI agent uses compute differently than a traditional application.

Over a single session, it writes and runs code, calls a model, and waits on a tool or a person before picking the work back up. As a result, it needs real compute in short, intensive bursts. It also needs a hard boundary around the code it just generated, so one agent’s mistake can’t spread to the host or to other customers.

This pattern is not unique to agents. The same pattern of heavy bursts, long idle waits, and strong isolation also applies to:

• Sandboxes for untrusted or generated code
• Per-user environments you create on demand and throw away
• Short-lived jobs such as CI/CD runners and preview environments

Traditional infrastructure does not fit these requirements. An always-on server running around the clock for a workload that is busy only for a few minutes an hour means paying for a lot of idle compute time. Containers start quickly and pack densely, but they share a kernel with their neighbors. This is more trust than you want to place in code an agent just wrote. You can close the gap yourself—with pausing, snapshots, fast startup, and per-tenant isolation—but that requires running your own virtualization platform. When your product is the agent platform, the virtualization layer underneath is undifferentiated work you’d rather not own.

Today, we are introducing DigitalOcean MicroVMs in private preview: isolated virtual machines (VMs) that pause when idle and resume exactly where they left off. MicroVMs are the layer that powers DigitalOcean Managed Agents. Now, they’re also available directly to teams building their own agent infrastructure.

DigitalOcean MicroVMs: one isolated VM per session, paused when idle

A MicroVM is a lightweight virtual machine that runs your container image in its own VM, with its own kernel, isolated from other workloads on the host. You bring a public or private image from DigitalOcean Container Registry, choose a size, and access the running workload over an authenticated HTTPS endpoint. DigitalOcean MicroVMs run on open-source Firecracker technology, so each VM has a real hardware-virtualized boundary rather than a shared kernel.

What differentiates a MicroVM from a traditional VM is how it behaves when the work stops. A MicroVM pauses automatically after a timeout you set and resumes on the next request with its memory, files, and running processes exactly as they were. It consumes compute only while it is running. Checkpoints handle the other half of the problem—the setup tax you would otherwise pay on every start. Boot a MicroVM once and let it load its runtime, dependencies, and models. Then checkpoint that warmed-up state and start new MicroVMs from it, instead of rebuilding the environment each time.

If you’re building agent infrastructure, you can treat this as a disposable machine. It starts fast, retains its state across a pause, and shuts down cleanly when the session ends. DigitalOcean runs the fleet, the isolation, and the lifecycle underneath.

MicroVMs use cases

MicroVMs are a clear fit for agent infrastructure, where every session needs its own isolated environment that starts on demand, pauses while it waits, and resumes with its state intact. The same properties apply to workloads beyond agents, especially any workload that creates many short-lived environments, runs them in bursts, and needs a hard boundary between them.

Here’s where MicroVMs fit:

• Coding and eval workspaces: A hardware-isolate MicroVM for each task, where an agent writes, runs, and tests code, then disposes of the environment when it is done.
• Long-running agent workflows: Agentic workflows that run for hours, pause while they wait on a model or a person, and resume exactly where they left off.
• Parallel reinforcement learning (RL) environments: Many identical environments running side by side and reset sub-second for the next episode.
• Ephemeral data per agent: A throwaway Postgres or scratch store for each agent to read, change, and discard. What it writes survives a pause and is cleared when the session ends.
• Per-tenant tools and MCP servers: Each customer’s tools run in their own VM, so one tenant can never reach another’s.
• Untrusted or generated code: Plugin sandboxes, security analysis, and evaluation jobs, with each run isolated in its own VM.
• Short-lived jobs: CI/CD runners, pull request (PR) preview environments, and ETL steps that run, finish, and shut down.

The common thread across these use cases is the need for many isolated environments, created on demand, busy in bursts, and cheap to discard.

MicroVMs: Already powering DigitalOcean Managed Agents

MicroVMs already power a demanding agent workload at scale: DigitalOcean Managed Agents (available in public preview) runs every agent session inside a MicroVM. Harness Runtime, the execution layer inside Managed Agents, relies on MicroVMs to isolate agent-generated code, pause a session the moment it goes idle, and resume it in under a second with files and state intact. Teams at OpenHands, Qencode, and Amplitude are already building on DigitalOcean Managed Agents.

MicroVMs expose the same primitive directly, and the two products meet different needs. Managed Agents is DigitalOcean’s hosted platform for running agents: you bring an agent, and we operate the runtime, tool access, and scaling underneath it. MicroVMs are the layer operating underneath, for teams who would rather build and control their own execution layer than adopt a managed one. Build your own agent platform, sandbox product, or code-execution service on MicroVMs, or anything that needs fast, isolated compute that pauses when idle.

DigitalOcean MicroVMs are currently available through an invite-only Private Preview. Request access.

• You can now prepay for Serverless Inference and Managed Agents with the Inference and Agents balance, which no other products draw from. Add $5 to $500 at a time or turn on auto-reload from the Billing page. Funds do not expire. You can also pay for eligible usage at a discount with a monthly plan, now in public preview. See Inference and Agents Balance and How to Pay for Harness Runtime.

• The DigitalOcean API now supports querying Insights metrics and logs and managing alert rules and notification channels in public preview. For available operations and endpoint details, see the DigitalOcean Insights API Reference.

• DigitalOcean Insights is now available in public preview, providing dashboards, metrics, logs, traces, and metric alerts for supported DigitalOcean resources and regions. For more information, see DigitalOcean Insights.

• The Agent Development Kit (ADK), including the gradient-adk Python package and the gradient CLI, is deprecated. You cannot deploy any new agents with ADK. Agents already deployed with ADK continue to run normally and you are billed for model usage. If you are using a DigitalOcean-hosted model, you are charged for those model keys. Migration guidance and end-of-support timeline will be published in the future.

Updates

• You can no longer install extensions on PostgreSQL Advanced Edition clusters with CREATE EXTENSION. The pre-installed pg_repack, pg_stat_statements, pgaudit, plpgsql, and vector extensions remain available. To request another extension, contact support. See Supported PostgreSQL Extensions.
• Organization owners and admins can now view and adjust resource limits for every team in their organization from the new Resource limits tab of the organization dashboard. The tab shows per-resource usage and limits for each team, supports inline edits within each team’s tier maximum, and lets you request increases beyond the maximum on a team’s behalf. For more information, see How to Manage Resource Limits for Organization Teams.
• The following OpenAI model is now available on DigitalOcean Inference for serverless inference and Agent Development Kit:
  • GPT-6.1 Sol

For more information, see the Available Models page.

• You can now set how long your DigitalOcean session remains active before you need to sign in again. On the My Account page, the Session duration setting accepts whole-number values from 1 hour to 30 days and applies to all browsers and devices where you’re signed in.

Isolated Worker Nodes for DigitalOcean Kubernetes (DOKS)

Isolated Worker Nodes for DigitalOcean Kubernetes (DOKS) are now in general availability. Every worker node in an isolated cluster runs without a public IPv4 address, so nodes are removed from the public internet at the network level rather than only protected by a firewall.

Outbound traffic, including node provisioning and container image pulls, routes through a VPC NAT Gateway, and other resources in the same VPC reach the nodes over private addresses. The Kubernetes API server stays publicly reachable so you can still manage the cluster. You can enable isolation only when you create a cluster running Kubernetes 1.36 or later.

v1.175.0 of doctl

v1.175.0 of doctl, the official DigitalOcean CLI, removes the doctl gradient command. The doctl gradient agent commands are deprecated with no CLI replacement. To manage agents, use the control panel or API. The doctl gradient knowledge-base commands are deprecated and will be removed in a future release. Use doctl knowledge-base instead. To list models, regions, and OpenAI keys, use the doctl serverless-inference commands.