AgentsCloud agent computers compared: every provider, status, price and region
✓ Last verified October 15, 2026
Pricing and features come from official pages. OSWorld measures agents on real computer tasks, but providers rarely publish scores for their own hosted product. Judge on status, admin controls, region and cost per task. Plain-English guide: /learn/cloud-agents. See the benchmarks we trust.
New to this? Read the plain-English guide first: Cloud agent computers: what they are and how to choose.
Cloud computers, then agents that borrow one per task
Cloud computers: the agent gets its own computer
A real cloud computer or desktop (a full VM) that is dedicated to your agents and that they operate themselves, with apps, a browser and files. Only 3 products qualify today.
| Product | Status | What it is and how it works | Price and plan | Where it runs | Security and admin controls |
|---|---|---|---|---|---|
| Grok Bot xAI | GA | Persistent agent "Bots" that work on a cloud computer with a browser, filesystem and terminal. Work keeps going while your laptop is closed, and Bots can run routines on a schedule. This is xAI's cloud computer agent. We found no separate computer-use tool in xAI's API docs. For: Individuals and teams who want always-on agent teammates. | Included on The X Subscription Premium ($30/mo), Super ($100/mo) and Ultra ($200/mo). Plus ($8/mo) doesn't include it. | The cloud computers run in Cursor's cloud. Apps are available for macOS, Windows, Linux, iOS and Android. | One cloud computer per account, shared by your Bots, with strict isolation between users. You sign in to sites yourself, and running anything on your own computer needs your approval. |
| Windows 365 for Agents Microsoft | GA | Microsoft's cloud PC product for agents: managed Windows Cloud PCs that an agent signs in to and operates like a person would, with full desktop apps and a browser. Agents check out a Cloud PC from a pool, or you can keep one always available. Microsoft says it powers Copilot Studio computer use, Researcher with Computer Use and Agent 365 agents. Generally available since June 2026. The agent to watch here is Autopilot in Microsoft Copilot (private preview, announced Sep 25, 2026, previously called Scout): a persistent agent that lives in your tenant with "its own identity, memory, computer and workspace" and that you @mention in Teams and Outlook. Microsoft hasn't said publicly whether Autopilot's computer is a Windows 365 for Agents Cloud PC, and it hasn't published an Autopilot price beyond "usage-based billing". Not to be confused with Windows Autopilot, the laptop-setup service. For: IT teams that need agents to use Windows apps and websites under company security policy. Autopilot is for Microsoft 365 organizations in its private preview. | Pay as you go: $0.40 per Cloud PC hour in the US, or $0.52 in Europe and the UK. An always-available Cloud PC adds $5/month in the US. Billed through Azure, with 50 free hours to start through Copilot Studio. | Azure regions in the US, Canada, Europe and the UK, Asia, India, Australia, South Africa, Brazil and the UAE. Prices vary by region. | Each Cloud PC is Entra-joined and enrolled in Intune. Agents get their own Entra agent identity and never use people's credentials. Conditional Access, Defender, Purview and an Intune security baseline all apply. |
| Manus Cloud Computer Manus | GA | An optional add-on to Manus: a persistent Ubuntu 24.04 server with a public IP that keeps bots, sites and databases running between sessions. (Ordinary Manus tasks without it run in a per-task cloud sandbox.) For: Non-technical users who want research, decks and always-on bots. | Manus plans run from Free to Pro ($20 or $40/mo). Cloud Computer costs $30/mo (2 vCPU, 4 GB) or $50/mo (2 vCPU, 8 GB). | Manus-managed cloud. | The Team plan adds a training opt-out. If you stop paying, the Cloud Computer and its files are deleted. |
Agents that borrow a cloud browser or sandbox per task
The difference: these agents also work in the cloud, but they borrow a browser, VM or code sandbox for one task and then it's reset or thrown away. There's no persistent computer of your own.
Cloud browser
Just a web browser in the cloud, lent to the agent for a task or conversation.
| Product | Status | What it is and how it works | Price and plan | Where it runs | Security and admin controls |
|---|---|---|---|---|---|
| ChatGPT Work: cloud browser OpenAI | GA | ChatGPT Work gets its own browser on a separate computer in the cloud. It reads pages, clicks, fills forms and signs in to sites through a secure sign-in form, then keeps going after you close your laptop. It replaced ChatGPT agent, which had already absorbed Operator; both are no longer available. For: Anyone on a paid ChatGPT plan who wants web errands and back-office web tasks done in the background. | Included on paid ChatGPT plans (not Free or Go). Rollout and workspace permissions can affect access. | OpenAI-managed cloud, in OpenAI's supported countries. The cloud browser keeps its own cookies and never uses your own browser's logins. | Per-site permissions (always ask, auto approve, always allow), allow and block lists, confirmation before payments and bookings, and a "take over" mode. Credentials typed into the sign-in form aren't visible to the model. |
| Researcher with Computer Use (Microsoft 365 Copilot) Microsoft | GA | Researcher can use a temporary sandbox computer to browse sites that need interaction. The sandbox runs on Windows 365 for Agents with Linux containers, is isolated from your device and intranet, and is deleted after the conversation. For: Microsoft 365 Copilot users doing cited research. | Needs a Microsoft 365 Copilot license ($30/user/mo, annual) and an admin to turn it on. | Within the Microsoft 365 data boundary. | Admins turn it on or off, limit it to pilot groups and set allowed or blocked domains. A network classifier checks every request, and no credentials are stored in the sandbox. |
Per-task cloud VM
The agent runs each task on a cloud machine it borrows, then hands it back. You never get a desktop of your own.
| Product | Status | What it is and how it works | Price and plan | Where it runs | Security and admin controls |
|---|---|---|---|---|---|
| Gemini Spark Google | Rolling out | Google's 24/7 personal agent. Tasks run on Google Cloud virtual machines, so they keep going after you close your device. It works with Gmail, Docs, Sheets, Slides, Keep and Tasks, can connect to apps such as Canva, Dropbox and OpenTable or any MCP server, and runs scheduled tasks. Project Mariner, Google's earlier browser agent, shut down on May 4, 2026. For: Google AI Ultra and US Google AI Pro subscribers. | Google AI Pro $19.99/mo (US, English) or Ultra from $99.99/mo. | Google Cloud. For Ultra, it's rolling out wherever Gemini Apps are supported except the EEA, the UK, Switzerland and Nigeria. For Pro, it's rolling out in the US. | Pauses for sign-ins and for approval of sensitive actions. This is a consumer Google AI plan feature. For company-wide agents with admin controls, Google sells Gemini Enterprise. |
| Copilot Studio computer use Microsoft | GA | A tool you add to a Copilot Studio agent so it can operate websites and Windows desktop apps. It runs on a hosted browser (Preview), on a Cloud PC pool from Windows 365 for Agents (Preview), or on your own machines. For: Makers automating legacy apps and websites that have no API. | Each step costs 5 Copilot Credits (15 with a premium model). Credits come in packs of 25,000 for $200/month, or pay as you go. Cloud PC pools are billed separately through Azure, with 50 free hours per tenant. | The hosted browser runs in a Microsoft-managed environment, and Cloud PC pools run in your chosen Windows 365 region. | Environment-level permissions and admin access controls. The hosted browser isn't joined to your tenant and Microsoft doesn't recommend it for production. Use a Cloud PC pool or your own machines for company resources. |
| Perplexity Computer and Comet Perplexity | GA | Perplexity Computer is a cloud "digital worker" that runs multi-step workflows across apps and the web. Comet, by contrast, is a browser you install on your own machine. For: Research-heavy professionals. | Computer is included with Pro and Max. Enterprise Pro starts at $40/seat/mo. | Perplexity-managed cloud for Computer. Comet runs locally. | The Enterprise Pro and Enterprise Max plans are sold per seat for organizations. Max adds Brain, a memory system that's in research preview. |
Code sandbox
A throwaway machine that clones your code, runs it and opens a pull request.
| Product | Status | What it is and how it works | Price and plan | Where it runs | Security and admin controls |
|---|---|---|---|---|---|
| Codex Cloud OpenAI | GA | Runs coding tasks on OpenAI-managed computers. You prepare an environment once (repos, tools, network access), and then each task runs in its own isolated workspace and keeps going while your computer sleeps. For: Developers and engineering teams already on ChatGPT. | Included with ChatGPT plans, with usage limits that grow by tier. | OpenAI-managed cloud. Saved VM state is recoverable for up to 7 days. | Admins control Codex cloud access in managed workspaces, and Enterprise can share environments. Codex in the cloud isn't covered by OpenAI's BAA, so don't use it for protected health information. |
| Claude Code cloud sessions Anthropic | GA | Claude Code runs in an isolated, Anthropic-managed VM (or your organization's self-hosted environment), clones your GitHub repo and works while you're away. Start it from the browser, the desktop or mobile app, or the terminal. Generally available since Sep 23, 2026. For: Developers and teams who want to hand off coding tasks in parallel. | Pro, Max and Team plans, and Enterprise premium or Chat + Claude Code seats. There's no separate charge for the cloud VM; it counts toward your plan's usage limits. | Anthropic-managed infrastructure by default, or a self-hosted environment. | Network access levels per environment, setup scripts, and "network secrets" that a proxy adds outside the VM so the agent never sees the key. Git goes through a secure proxy. Owners can share standard environments with the organization. |
| Jules Google | Preview | Google's asynchronous coding agent, which Google labels beta. It clones your GitHub repo to a cloud VM, proposes a plan, shows a diff and opens a pull request. For: Developers on GitHub. | Free 15 tasks/day; AI Pro 100/day; Ultra 300/day. | Google-managed cloud VM. | You approve the plan and the pull request. It works only on the repos you connect. |
| GitHub Copilot cloud agent GitHub (Microsoft) | GA | Assign a GitHub issue to Copilot and it works in a cloud environment, then opens a pull request. For: Teams whose code lives on GitHub. | Uses GitHub AI Credits from your Copilot plan: Pro $10/mo, Business $19/seat/mo, Enterprise $39/seat/mo. | GitHub-hosted. | Organization policies for Copilot, and every change goes through pull request review. |
| Cursor Cloud Agents Cursor | GA | Coding agents that run in dedicated, isolated cloud VMs with snapshots, secrets and network controls. For: Developers using Cursor. | Premium plan ($30/mo) and above. Usage is billed at the model's API rate, with no separate VM charge. | Cursor-managed cloud. | Secrets, network egress controls and team admin settings. |
| Devin Cognition | GA | An autonomous software engineer that takes tickets and works on its own cloud machine. For: Engineering teams with well-scoped backlogs. | Free; Pro $20/mo; Max $200/mo; Teams $80/mo plus $40/mo per dev seat; Enterprise custom. | Cognition-managed cloud. | Team and Enterprise admin controls. |
Infrastructure for builders
Browsers, sandboxes and desktops your developers plug into agents they build themselves.
| Product | Status | What it is and how it works | Price and plan | Where it runs | Security and admin controls |
|---|---|---|---|---|---|
| Claude computer use (API) Anthropic | GA | A developer tool: Claude sees screenshots and sends mouse and keyboard actions, and your own app carries them out on a VM or container you run. The current toolset (computer_toolset_20260801) needs no beta header. For: Developers building their own computer-using agents. | Standard API token pricing. The toolset adds about 4,500 input tokens per request, and screenshots are billed as image input. | Claude API and Google Cloud. The desktop itself runs wherever you host it. | You control the environment. Screenshots and actions stay in your environment, and it's eligible for zero data retention. Anthropic scans screenshots for prompt injection. |
| Amazon Bedrock AgentCore Browser and Code Interpreter, and Nova Act Amazon (AWS) | GA | AgentCore is a managed cloud browser and code sandbox that your own agents call per session, with session isolation, Live View and Session Replay. Nova Act is an AWS service and SDK for agents that automate browser workflows in production. For: Developers building agents on AWS. | AgentCore: $0.0895 per vCPU-hour and $0.00945 per GB-hour, billed per second; idle time during I/O waits is free. Nova Act: $4.75 per agent hour; time spent waiting for a human isn't billed. | AgentCore: AWS Regions where it is offered. Nova Act: US East (N. Virginia). | IAM, isolated sessions, AgentCore Identity and Policy, and CloudWatch observability. |
| E2B E2B | GA | Open-source cloud sandboxes for running agent code. Sandboxes are CPU-only, with 1 to 8 vCPUs. For: Developers adding a safe "computer" to their own agent. | Hobby $0 with $100 in one-time credits; Pro $150/mo plus usage. Usage costs $0.000014 per vCPU-second and $0.0000045 per GiB-second. Enterprise has a $3,000/mo minimum. | E2B cloud. Bring-your-own-cloud is available on Enterprise. | Session limits of 1 hour on Hobby and 24 hours on Pro, plus concurrency caps. |
| Browserbase Browserbase | GA | Fleets of cloud browsers for agents, plus the open-source Stagehand SDK. For: Developers building browser agents. | Free (1 browser hour); Developer $20/mo (100 hours, then $0.12/hour); Startup $99/mo (500 hours, then $0.10/hour); Scale custom. | US East, US West, EU Central and UK, with Asia Pacific on request. | SOC 2, with HIPAA BAA, DPA and SSO on Scale. Data retention is 7 to 30+ days depending on plan. |
| Cua Spaces Cua | GA | Shared desktops for people and agents on computers you already own or rent, such as a Mac, a Linux box or a VM in your own cloud. Cua doesn't host the machines. For: Teams that want agents and people working on the same desktop. | Free; Pro $18/mo; Teams $30/seat/mo. Your cloud provider bills you for any VMs. | Your own machines, or your AWS, Google Cloud or Modal account. | The Teams plan adds roles and device approval. Cua says SSO and audit export aren't included yet. |
Checked and not listed
- OpenAI Operator and ChatGPT agent: Retired. Operator moved into ChatGPT agent, which was then replaced by the cloud browser in ChatGPT Work. Source
- Google Project Mariner: Shut down on May 4, 2026. Google now points people to Gemini Spark for complex tasks. Source
- Scrapybara: Its site says it "built computers for agents" and the team now builds a different product, so we no longer list it. Source
Status, prices and regions are from each vendor's official page, checked 2026-10-15. "Preview" means the vendor labels it preview or experimental. "Rolling out" means the vendor is releasing it in stages.
Cloud-hosted automation (not a cloud computer)
n8n Cloud, Zapier and Make run agent workflows on their servers (trigger → AI or app step → action). They do not give the agent its own desktop or browser. Full plain-English guide: Cloud-hosted automation.
| Product | How it works | AI / agent features | Price (checked 2026-10-15) | Self-host? |
|---|---|---|---|---|
| n8n Cloud n8n | You build a visual workflow: a trigger (new email, form, schedule), then steps that call apps, run code or ask an AI node, then an action (Slack message, CRM update). n8n Cloud hosts the instance so you do not manage servers. One run of the whole workflow counts as one execution, no matter how many steps it has. | AI Agent node, MCP client/server, and other AI nodes are included on Cloud plans. Assistant (preview) helps you build workflows and comes with monthly Assistant credits (1,600 on Starter; up to 9,600 on Pro). You can also use AI models through n8n without bringing your own API keys. | Cloud Starter $24/mo monthly or $20/mo billed annually (2.5K executions). Pro $60/mo monthly or $50/mo billed annually (10K executions). Business $960/mo monthly or $800/mo billed annually (40K executions; self-hosted license). Enterprise custom. Unlimited users and workflows on every plan. | Yes. The Community Edition is free on GitHub and includes AI agents and workflows. Paid Business and Enterprise licenses unlock SSO, Git version control, queue mode and other team features on your own servers. |
| Zapier (AI by Zapier / Agents) Zapier | You connect apps and build a Zap: a trigger, then one or more action steps. AI by Zapier is an AI step inside the Zap editor that can reason, call tools and pass results to the next step. Zapier Agents (agents.zapier.com) still exists; Zapier is migrating Agents into AI by Zapier so agent runs live in the same editor and history as other Zaps. | AI by Zapier steps support model tiers (Standard, Advanced, Premium), per-tool approvals and structured outputs. Copilot helps build Zaps from plain language. Agents that have not migrated yet still run at agents.zapier.com and are measured in activities on the Agents add-on. | Free $0 (100 tasks/mo). Professional from $19.99/mo. Team from $69/mo. Enterprise custom. AI by Zapier uses the same task pool as other Zap steps (no separate Agents subscription once migrated). Higher task tiers cost less per task. Zapier has not set a shut-off date for Agents yet. | No. Zapier runs only as a hosted cloud service. |
| Make Make | You build a visual "scenario": modules for triggers, filters, routers and actions across 3,500+ apps. Each module that processes data uses credits (most use 1 credit). Make hosts the scenarios on its cloud. There is no separate desktop or browser for the agent to look at and click. | Make AI Agents (open beta) are available on all plans with Make's AI Provider, or with your own OpenAI/Anthropic key on paid plans. Running an agent costs 1 credit per operation plus token-based credits when you use Make's AI Provider. Make AI Toolkit and other AI modules are also available. | Free $0 (1,000 credits/mo). At 10,000 credits/mo: Core $12/mo monthly (from $9/mo billed annually), Pro $21/mo monthly (from $16/mo billed annually), Teams $38/mo monthly (from $29/mo billed annually). Enterprise custom. Extra credits can be purchased in bundles. | No public self-host edition. Make is a hosted cloud product. Enterprise can add on-premises agents to reach local networks. |
Related agent profiles on this site: n8n, Zapier Agents and Lindy. Lindy is a no-code AI assistant for inbox and meetings, not a step-based workflow tool, but readers often compare it with these.
Quick verdict
OpenAI
ChatGPT Work, Codex & Agents SDK
Choose it if: everyday users who already pay for chatgpt.
Anthropic
Claude Code
Choose it if: developers who live in the terminal.
Google
Gemini, Jules & Gemini Enterprise
Choose it if: google workspace teams.
xAI
Grok & Grok Bot
Choose it if: people who want always-on agents for real projects.
Microsoft
Windows 365 for Agents
Choose it if: it teams that need agents to use windows apps under company policy.
Microsoft
Microsoft Copilot Studio
Choose it if: microsoft-centric organizations building internal or customer-facing agents.
GitHub (Microsoft)
GitHub Copilot (agent mode, cloud agent & CLI)
Choose it if: teams whose code already lives on github.
Amazon (AWS)
Amazon Bedrock AgentCore & Nova Act
Choose it if: teams already building on aws.
Manus
Manus
Choose it if: research reports and decks.
Perplexity
Perplexity & Comet
Choose it if: fast, cited research.
Cursor
Cursor
Choose it if: developers who want ai inside their editor.
Cognition
Devin & Devin Desktop (formerly Windsurf)
Choose it if: engineering teams with a backlog of well-defined tickets.
E2B
E2B
Choose it if: developers adding code execution to an agent.
Browserbase
Browserbase
Choose it if: developers building browser agents.
Side by side
| Agent | Pricing | Best for | Privacy | Skill |
|---|---|---|---|---|
| ChatGPT Work, Codex & Agents SDK OpenAI | The cloud browser in ChatGPT Work is on paid plans (not Free or Go). Codex is included in ChatGPT Free, Go, Plus, Pro, Business and Enterprise, with usage that grows by tier. | Everyday users who already pay for ChatGPT; Delegating web research and form-filling; Coding tasks in the cloud | Cloud. Business, Enterprise and Edu plans have separate data controls. | Beginner |
| Claude Code Anthropic | Included with Claude Pro ($20/mo, or $17/mo billed annually) and Max (from $100/mo), and shares the plan's usage limits. Team Standard seats are $20/seat/mo annually ($25 monthly). You can also pay per token through the API. | Developers who live in the terminal; Repo-wide changes with tests | Cloud model; your code stays on your machine except what's sent as context. Also available through Amazon Bedrock and Google Cloud Vertex AI. | Intermediate |
| Gemini, Jules & Gemini Enterprise Google | Google AI Pro $19.99/mo; Ultra $99.99/mo (5x Pro usage) or $199.99/mo (20x). Gemini Enterprise Business $14/user/mo, Standard from $21/user/mo. | Google Workspace teams; Parallel background coding tasks; Companies wanting a governed agent catalog | Cloud. Enterprise editions add admin governance. | Beginner |
| Grok & Grok Bot xAI | The X Subscription: Free; Plus $8/mo; Premium $30/mo; Super $100/mo; Ultra $200/mo (annual discounts available). Plus doesn't include Grok Bot. | People who want always-on agents for real projects; Cursor users who also use Grok | Cloud. | Intermediate |
| Windows 365 for Agents Microsoft | Pay as you go: $0.40 per Cloud PC hour in the US, $0.52 in Europe and the UK, and $0.80 in South America. An always-available Cloud PC adds $5/month in the US. Billed through Azure, with 50 free hours to start through Copilot Studio. | IT teams that need agents to use Windows apps under company policy; Automating legacy desktop apps with no API | Runs in your chosen Azure region in the US, Canada, Europe and the UK, Asia, India, Australia, Africa, Brazil or the UAE. | Advanced |
| Microsoft Copilot Studio Microsoft | Copilot Credit packs: 25,000 credits for $200/mo (tenant-wide), or pay-as-you-go. Prepaying with Commit Units saves up to 20%. Microsoft 365 Copilot users can build agents for internal use, but publishing to outside channels needs a standalone Copilot Studio plan. | Microsoft-centric organizations building internal or customer-facing agents | Cloud, in your Microsoft tenant, with data-residency options. | Intermediate |
| GitHub Copilot (agent mode, cloud agent & CLI) GitHub (Microsoft) | Free $0 (limited chat and agent use); Pro $10/mo; Pro+ $39/mo; Max $100/mo; Business $19/seat/mo; Enterprise $39/seat/mo. Chat, agents, code review and CLI use GitHub AI Credits (1 credit = $0.01), and each plan includes a monthly allowance. Code completions don't use credits on paid plans. | Teams whose code already lives on GitHub; Developers who want one assistant across many editors; Assigning issues to an agent that opens pull requests | Cloud. GitHub says it doesn't train on Business or Enterprise data. Free, Pro and Pro+ interactions may be used for training unless you opt out in settings. | Beginner |
| Amazon Bedrock AgentCore & Nova Act Amazon (AWS) | AgentCore Browser and Code Interpreter: $0.0895 per vCPU-hour and $0.00945 per GB-hour, billed per second, and idle time during I/O waits is free. Nova Act: $4.75 per agent hour. | Teams already building on AWS; Production browser automation at scale | Runs in your AWS account's chosen Region, with IAM controls. Nova Act runs in US East (N. Virginia). | Advanced |
| Manus Manus | Free (300 daily credits, Lite model); Pro from $20/mo or $40/mo; Team from $20/seat/mo. Annual billing saves 17%. | Research reports and decks; Recurring scheduled tasks | Cloud. Team plan adds a data-training opt-out. | Beginner |
| Perplexity & Comet Perplexity | Check perplexity.ai for current plans. We couldn't verify today's tier prices on an official page. | Fast, cited research; Browser-based tasks | Cloud. | Beginner |
| Cursor Cursor | Free; Plus $8/mo; Premium $30/mo; Super $100/mo; Ultra $200/mo; Teams $30, $100 or $200 per seat. Legacy Pro ($20) plans continue for existing subscribers. | Developers who want AI inside their editor; Teams standardizing on one coding tool | Cloud; enterprise admin controls. | Intermediate |
| Devin & Devin Desktop (formerly Windsurf) Cognition | Free $0; Pro $20/mo; Max $200/mo; Teams $80/mo plus $40/mo per full dev seat (up to 200 users); Enterprise custom. Existing Windsurf plans carried over unchanged. | Engineering teams with a backlog of well-defined tickets | Cloud. | Advanced |
| E2B E2B | Hobby $0 with $100 in one-time credits; Pro $150/mo plus usage. Usage costs $0.000014 per vCPU-second and $0.0000045 per GiB-second, and storage is included. Enterprise has a $3,000/mo minimum. | Developers adding code execution to an agent | E2B cloud. Bring-your-own-cloud is available on Enterprise. | Advanced |
| Browserbase Browserbase | Free (1 browser hour); Developer $20/mo (100 hours, then $0.12/hour); Startup $99/mo (500 hours, then $0.10/hour); Scale custom. | Developers building browser agents; Back-office web automation | US East, US West, EU Central and UK. SOC 2, with HIPAA BAA and SSO on Scale. | Advanced |
Sources (official pages, checked October 15, 2026)
- ChatGPT Work, Codex & Agents SDK: Using cloud browser in ChatGPT
- ChatGPT Work, Codex & Agents SDK: ChatGPT agent (retired)
- ChatGPT Work, Codex & Agents SDK: ChatGPT Atlas
- ChatGPT Work, Codex & Agents SDK: Introducing ChatGPT Atlas
- ChatGPT Work, Codex & Agents SDK: Codex pricing
- ChatGPT Work, Codex & Agents SDK: ChatGPT plans
- ChatGPT Work, Codex & Agents SDK: AgentKit post (with wind-down update)
- ChatGPT Work, Codex & Agents SDK: API pricing
- Claude Code: Claude Code
- Claude Code: Claude plans & pricing
- Gemini, Jules & Gemini Enterprise: Jules limits and plans
- Gemini, Jules & Gemini Enterprise: What's new for Gemini Spark
- Gemini, Jules & Gemini Enterprise: Gemini in Chrome auto browse
- Gemini, Jules & Gemini Enterprise: Google AI plans
- Gemini, Jules & Gemini Enterprise: Gemini for Work pricing
- Gemini, Jules & Gemini Enterprise: Gemini Enterprise editions
- Grok & Grok Bot: xAI docs: X Subscription pricing
- Grok & Grok Bot: xAI: Grok Bot on more plans
- Grok & Grok Bot: xAI: DeepSearch (Grok 3 launch)
- Windows 365 for Agents: What is Windows 365 for Agents?
- Windows 365 for Agents: Pricing by region
- Windows 365 for Agents: Billing and regions
- Windows 365 for Agents: Identity and security
- Microsoft Copilot Studio: Copilot Studio pricing
- Microsoft Copilot Studio: Copilot Studio product page
- GitHub Copilot (agent mode, cloud agent & CLI): Copilot plans & pricing
- GitHub Copilot (agent mode, cloud agent & CLI): GitHub Docs: Copilot plans (incl. Business & Enterprise)
- GitHub Copilot (agent mode, cloud agent & CLI): GitHub Docs: Copilot cloud agent
- GitHub Copilot (agent mode, cloud agent & CLI): GitHub Docs: Copilot CLI
- GitHub Copilot (agent mode, cloud agent & CLI): VS Code: Copilot chat and agents
- Amazon Bedrock AgentCore & Nova Act: AgentCore pricing
- Amazon Bedrock AgentCore & Nova Act: What is Nova Act?
- Amazon Bedrock AgentCore & Nova Act: Amazon Nova pricing
- Manus: Manus membership pricing
- Perplexity & Comet: Perplexity Comet
- Cursor: Cursor pricing and plans
- Devin & Devin Desktop (formerly Windsurf): Devin pricing
- Devin & Devin Desktop (formerly Windsurf): Windsurf is now Devin Desktop
- Devin & Devin Desktop (formerly Windsurf): Devin Desktop FAQ
- E2B: E2B pricing
- Browserbase: Browserbase pricing
Benchmarks: SWE-bench Verified · OSWorld · GAIA · HAL (Princeton) · Artificial Analysis