Alternatives
Devin Alternatives & Competitors (2026): 8 Best Options
What are the best Devin alternatives?
When a company with five to twenty engineers already pays Anthropic or OpenAI, the best Devin alternative is Jiro: it hosts Claude Code, Codex or Cursor Agent using the login or API key you hold, charges a flat runtime fee, and adds no token markup. Jiro is a background coding agent by Nori. The runners-up, in order, are Cursor's cloud agents, Codex cloud from OpenAI, Anthropic's hosted Claude Code sessions, Factory, the Copilot cloud agent from GitHub, Tembo and OpenHands. It is not Amazon's Kiro or Atlassian's Jira.
Devin and eight alternatives side by side
| Model supplier | Billing | Runs where | Starts from | Ships as | |
|---|---|---|---|---|---|
| Jiro | Your Claude/ChatGPT login or API key; Cursor key [1] | Free 30-day trial; Developer $99/mo; Team $250/mo [2] | Isolated cloud environment with your repos and tools [3] | Slack, web, Discord, terminal, schedules, webhooks [4] | PRs, commits, comments for review [5] |
| Devin | Anthropic, OpenAI, Google, Cognition SWE; Adaptive routing [6] | Free; Pro $20/mo; Max $200/mo; Teams $40/seat, $80/mo min [7] | Cognition cloud, or local CLI/Desktop [8] | GitHub/GitLab/Bitbucket/ADO; Slack, Teams, Linear, Jira [9] | Opens PRs and answers PR comments [10] |
| Cursor cloud agents | Plan usage at the selected model's rate [11] | Premium $30/mo+ (X Subscription Corp); third-party at API price [12] | Cursor cloud VMs or self-hosted machines [13] | Desktop, web, mobile, Slack, GitHub, Linear, API [14] | PRs with screenshots and video [15] |
| OpenAI Codex cloud | OpenAI models; cloud needs a ChatGPT plan [16] | In ChatGPT Plus $20/mo and up [17] | Codex-managed cloud VMs [18] | ChatGPT web, mobile, desktop, CLI [19] | You open the PR [20] |
| Claude Code on the web | Claude models; Anthropic account required [21] | In Pro $20/mo, Max $100, Ultra $200, Team seats [22] | Anthropic-managed VMs [23] | Web, mobile, desktop, terminal [24] | You create the PR; auto-fix follows up [25] |
| Factory | Multi-model; own keys in CLI/desktop only [26] | Pro $20/mo; Plus $100; Max $200; Teams $60 + $40/seat [27] | Local, Factory cloud, your machines or cloud [28] | App, CLI, IDE, Slack, Teams, Linear, Jira [29] | Opens PRs from delegations [30] |
| GitHub Copilot cloud agent | Claude, Gemini, GPT, MAI via GitHub; no own keys [31] | Pro $10; Pro+ $39; Max $100; Business $19/user [32] | Cloud sandbox, Actions (59 min) or remote env [33] | GitHub, IDEs, CLI, API, Slack, Teams, Linear, Jira [34] | Draft PRs; cannot merge [35] |
| Tembo | Multiple harnesses; own keys or subscriptions [36] | Pro $60/mo ($90 usage); Max $200/mo ($400); VMs/sec [37] | Tembo cloud sessions (Docker or VM) [38] | GitHub, Slack, Linear, Jira, Sentry, schedules [39] | Opens PRs, including multi-repo PRs [40] |
| OpenHands | Own agent or Claude Code, Codex, Gemini, Pi; any LLM [41] | Free local; Teams $500/mo + $50/seat; Enterprise custom [42] | Local by default; optional OpenHands Cloud/Enterprise [43] | Web UI, API, GitHub, GitLab, Bitbucket, Slack [44] | Label or @-mention on GitHub [45] |
Pricing and limits with units (checked 2026-11-06)
| Product and plan | Price per unit | Included quota | Overage | Team cap and concurrency |
|---|---|---|---|---|
| Jiro Developer | $99/mo, one user [2] | Up to three persistent runtimes [2] | None from Nori; model usage stays on your own plan or key [46] | One user [2] |
| Jiro Team | $250/mo [2] | Five runtimes shared by the team [2] | None from Nori; zero token markup [46] | Multiple users; five runtimes, which sleep when idle [47] |
| Jiro trial | Free for 30 days, no card [2] | Five runtimes [47] | None from Nori [46] | Five runtimes for 30 days [47] |
| Devin Free | Free [7] | Limited usage plus Devin Review and DeepWiki [48] | On-demand credits [48] | Single user; up to 10 concurrent sessions [49] |
| Devin Pro | $20/mo [7] | Daily and weekly quota [48] | On-demand credits that never expire [48] | Single user; up to 10 concurrent sessions [49] |
| Devin Max | $200/mo [7] | Larger weekly quota, no daily cap [48] | On-demand credits that never expire [48] | Single user; no concurrency cap [49] |
| Devin Teams | $40/mo per full seat, $80/mo minimum [7] | Pro-sized quota per full seat; flex seats are free and draw on shared credits [50] | Shared on-demand credit pool; Adaptive meters at $0.50 in and $2.00 out per million tokens [51] | Up to 200 members [50]; no concurrency cap [49] |
| Devin Enterprise | Custom contract, metered in ACUs [52] | Per contract [52] | ACUs [52] | Dedicated single-tenant deployment available [53] |
| Cursor Premium | $30/mo [12] | Plan usage at the selected model's rate [11] | Third-party models at API list price [12] | Not stated on the pricing page |
| ChatGPT Plus (Codex cloud) | $20/mo [17] | Usage limits shared with ChatGPT [17] | Not stated on the pricing page | Not stated on the pricing page |
| ChatGPT Business (Codex cloud) | $25/mo per user, or $20/mo billed annually [17] | Usage limits shared with ChatGPT [17] | Not stated on the pricing page | Not stated on the pricing page |
| Claude Pro (Claude Code on the web) | $20/mo [22] | Plan usage [22] | Usage credits at API rates [54] | Not stated on the pricing page |
| Claude Max (Claude Code on the web) | $100/mo [22] | Five times Pro's usage [55] | Usage credits at API rates [54] | Not stated on the pricing page |
| Factory Teams | $60/mo plus $40/mo per seat [27] | Not stated on the pricing page | Not stated on the pricing page | Up to 10 seats [27] |
| GitHub Copilot Business | $19/mo per user [32] | Not stated on the pricing page | AI Credits at per-model rates, one credit = $0.01 [56] | Actions sessions capped at 59 minutes [33] |
| Tembo Pro | $60/mo [37] | $90 of usage a month [37] | Cloud machines metered by the second, from $0.1066 an hour [37] | Up to 5 users [57] |
| Tembo Max | $200/mo [37] | $400 of usage a month [37] | Cloud machines metered by the second, from $0.1066 an hour [37] | Up to 10 users [57] |
| OpenHands Teams | $500/mo [42] | 5,000 sessions a month [42] | $50/mo per extra seat [42] | 10 users included [42] |
What does Devin cost in 2026, and why do teams look for alternatives?
Devin's sticker prices are modest. Pro costs $20 monthly, Max $200, and the Teams plan is $40 a month per full seat with an $80 monthly minimum. [7] Teams go shopping because of the second line on the invoice. Every plan carries a quota, daily and weekly on Pro and Teams, weekly only on Max, and once it is gone the meter switches to on-demand credits. [48] With the Adaptive router switched on, self-serve quota drains at a flat $0.50 per million input tokens and $2.00 per million output tokens, whichever model Cognition chose for that step. [51]
Nothing wrong with that as a business. The trouble is that it is a second model bill for a company that already pays Anthropic or OpenAI, because Cognition is reselling the tokens. One narrow exception exists: a Pro, Max or Teams user may link a personal ChatGPT plan, and GPT requests then bill there, while Claude and Gemini traffic plus the machine time stay on Devin quota. [58] The linking facility is documented for ChatGPT plans only, so Claude usage keeps drawing on Devin quota. [58]
The product also keeps changing shape. The ACU-based Core and Team plans were retired on 2026-04-14. [52] Windsurf was folded into Devin Desktop in June. [59] Knowledge and playbooks are both deprecated in favour of Skills. [60] [61] Devin is not a bad product. It is one that needs a yearly re-evaluation, and this is the year.
How did we rank these eight Devin alternatives?
Four questions, weighted in this order. Who bills for tokens: the agent vendor, or the model company your finance team already pays? Jiro accepts the Claude or ChatGPT login your engineers use today, or an API key, and marks none of it up. [1] [46] Second: where does the agent execute, and does that place contain your toolchain or just a bare shell? [3] Third: how does a teammate hand it a task, and can a schedule or webhook do the same without a person present? [62] Fourth: is the deliverable a reviewable pull request, or a branch somebody must still finish? [5]
The ranking assumes a GitHub-hosted codebase, a handful to twenty engineers, and at least one frontier-model subscription on the company card. A four-hundred-person organisation on GitLab with a procurement department would sort this list differently, and it should.
1. Jiro: the Devin alternative when the model bill already exists
Jiro is the pick when your company already pays for Claude or ChatGPT. You choose the harness: Codex, Claude Code, Cursor Agent, or the OpenCode and pi-acp agents. [63] The first two sign in with the subscription or key your team already uses; Cursor Agent wants a Cursor API key. [1] Nori's price list has no line for tokens. [46]
What you pay Nori for is runtimes. Developer, at $99 monthly, covers one user with as many as three; Team, at $250 monthly, gives a whole team five to share; the 30-day trial asks for no card. [2] Idle runtimes sleep and wake on demand. [47] Every session has an isolated cloud environment of its own, prepared by your organisation's setup script, and no machine serves two sessions at once. [3] [64] [65]
Hand it work with a Slack mention, in the dashboard, from Discord or Handroll, or let a cron schedule or webhook do it; it answers with a PR, a commit or a comment. [4] [5] Miravoice now merges over 25% of its pull requests from Jiro's work. [66]
2. Cursor cloud agents: best if the whole team already lives in Cursor
Cursor rebranded its background agents as Cloud Agents. [67] They execute in Cursor-managed cloud machines, or you can move tool execution to hardware you manage. [13] Start one from the editor, from a phone, from a Slack message or a Linear ticket. [14] The resulting PR carries screenshots and a video of what the agent did. [15]
Billing is the catch. Cloud agents need Cursor's Premium tier, $30 a month, and any non-Cursor model is billed on top at the API's list rate. [12] Usage comes out of your Cursor plan at the chosen model's rate, so a Claude Max or ChatGPT subscription contributes nothing. [11] For a Cursor-native team that is acceptable. For everyone else it is Devin's economics under a different logo.
3. OpenAI Codex cloud: the cheapest route for ChatGPT-only teams
Codex cloud is part of ChatGPT Plus, so a $20 monthly seat already includes it, and higher tiers scale from there. [17] Tasks run on OpenAI's cloud machines while your laptop is closed. [18] You start them from the ChatGPT apps or the Codex CLI. [19]
Two constraints decide whether this works for you. The models are OpenAI's only, and holding an API key does not open the cloud side; a ChatGPT plan does. [16] And the pull request is yours to open once you have read the diff. [20] In Slack the bot is now addressed as @ChatGPT. [68] If the team is ChatGPT-only and content to finish PRs by hand, nothing on this page costs less. Teams that want Claude in the loop should keep reading.
4. Claude Code on the web: best for individuals on a Claude plan
Anthropic's hosted sessions ride along with a Pro, Max, Ultra or Team seat; the cloud machine itself is not billed separately. [22] Each session boots a fresh Anthropic-managed machine with 16 GB of memory, or runs in a self-hosted environment if your organisation routes it there. [23] You can launch one from claude.ai/code, from the apps, or with claude --cloud. [24]
It is Claude only, deliberately, and needs an Anthropic account rather than Bedrock or Google Cloud. [21] Sessions push a branch and leave the PR to you, though an auto-fix feature can later patch CI failures and respond to review comments. [25] Shared Slack work on Team and Enterprise has moved over to Claude Tag. [69] As a personal tool for a Claude subscriber it is hard to beat on price. As a team's background agent it lacks a shared environment and shared triggers.
5. Factory: best if you want an entire platform, not just an agent
Factory's Droids will run almost anywhere: on your laptop, on Factory's hosted machines, on a computer you remote-control, or inside your own cloud. [28] Delegations arrive from chat and ticket tools, Microsoft Teams included, and a delegated Linear issue comes back as a pull request with its URL. [29] [30] Prebuilt automation starters turn tickets and alerts into PRs on a schedule or an event. [70]
Factory's Teams plan is $60 monthly plus $40 for each seat, capped at ten seats; individuals pay $20, $100 or $200. [27] Bringing your own keys is allowed in the CLI and desktop app, but the hosted web and mobile apps will not take them. [26] A VPC option, Factory Private, remains in private preview. [71] It is the closest thing to Devin's breadth on this list, and the most moving parts for a team of eight.
6. GitHub Copilot cloud agent: the default if procurement already bought Copilot
GitHub's current name for it is Copilot in the cloud. The agent researches the repo, drafts a plan and carries it out in a GitHub-run sandbox, an Actions job or a remote environment of yours. [72] Sessions can be kicked off from GitHub.com, the IDEs, the CLI or API, and from chat and ticketing tools. [34] If the session runs on Actions, it is cut off at 59 minutes. [33]
Business costs $19 per user monthly and Pro $10, and since 2026-06-01 every plan meters tokens in AI Credits worth one cent each. [32] [56] The agent picks from Claude, Gemini, GPT and MAI models that GitHub supplies; GitHub documents BYOK for the editors and CLI only, not for this agent. [31] It opens draft PRs on a branch of its own and is barred from approving or merging them. [35] Least paperwork, least control.
7. Tembo: a multi-harness cousin with metered compute
Tembo is philosophically the nearest thing to Jiro on this list. It picks a harness per task, Claude Code or Codex among them, and will use your own Anthropic or OpenAI key or an existing Claude or ChatGPT subscription. [36] Sessions get an isolated container or machine with a live desktop. [38] Triggers include GitHub, Slack, Linear, Sentry and PagerDuty, as well as webhooks and schedules. [39]
The meter is where it diverges. Pro costs $60 monthly and bundles $90 of usage; the $200 Max tier bundles $400; and cloud machine time is billed per second on top. [37] Jiro's runtime fee does not move with the clock. Which is cheaper depends on how many machine-hours your team burns, so model a real month before choosing between the two.
8. OpenHands: best if you want the source code and will run it yourself
OpenHands is open source under the MIT licence and bills itself as a control center for coding agents. [73] Local is the default, with hosted OpenHands Cloud and Enterprise tiers as options. [43] It drives its own agent or several others, Claude Code and Codex included, against whichever LLM you point it at. [41] On GitHub, a label or an @openhands comment starts a run that may end in a PR. [45]
Running Agent Canvas yourself, alone, costs nothing. The Teams tier is $500 monthly for as many as ten people, with further seats at $50 and a ceiling of 5,000 sessions a month. [42] Somebody on your side becomes the operator. That is both the appeal and the cost, and it is why OpenHands sits eighth for a team with no platform engineer.
Which Devin alternatives accept a Claude or ChatGPT subscription directly?
Only two on this list take the subscription itself rather than a card. Jiro accepts a Claude OAuth login or setup token for Claude Code and Codex OAuth for Codex, alongside plain API keys. [1] Tembo accepts a Claude, ChatGPT or SuperGrok subscription as well as your own keys. [36] Anthropic's hosted Claude Code is, by definition, your Claude plan, and Codex cloud is your ChatGPT plan, but each is locked to its own vendor's models. [21] [16]
The rest resell. Cursor bills cloud-agent usage against its own plan; GitHub meters AI Credits; Factory takes your keys only in its local apps; Devin links a personal ChatGPT plan for GPT traffic and nothing else. [11] [56] [26] [58] For a Claude subscriber the mechanics on Jiro are a single command: claude setup-token, run on your own laptop, yields a token good for a year and tied to your plan tier. [74] No equivalent exists on Devin.
Which Devin alternative costs least when Claude or ChatGPT is already paid for?
Jiro, in nearly every configuration, because its only recurring charge is the runtime fee. Team is $250 monthly, the five runtimes are shared, and tokens flow from whatever Anthropic or OpenAI plan the company already holds, with no markup from Nori. [2] [46]
Devin Teams, by comparison, is $40 a month per full seat with an $80 monthly minimum, and then credits for whatever the quota does not cover. [7] [50] Those credits are the whole model bill, paid a second time. Watch Codex cloud, though. Inside ChatGPT Plus at $20 a month it undercuts everything here, provided OpenAI-only models and hand-finished pull requests suit you. [17] [20]
No model subscription yet? Add one to Jiro's side before comparing. Claude Max is $100 monthly per person, and Anthropic's own figures put heavy Claude Code use near $150 to $250 per developer per month on the API. [55] [75]
When Devin is the better choice
Devin wins on reach. Beyond Slack it listens in Microsoft Teams, in both major ticket trackers, and in PagerDuty and Pylon, and it works with all four big git hosts, Azure DevOps included. [9] Its Automations start sessions from events in those tools as well as from schedules and webhooks. [76] Jiro's sourced entry points are Slack and Discord, the web dashboard and the terminal, plus schedules and webhooks, and its documented repo connection is GitHub. [4] [77] If GitLab or Teams is load-bearing for you, Devin is the answer and this list is not.
Devin wins on model routing too. Adaptive chooses a model per step from Cognition's own SWE family alongside Anthropic, OpenAI and Google, so nobody on the team has to decide. [6] Jiro makes you pick a harness and a credential.
Devin Desktop is a real IDE descended from Windsurf, and other ACP-speaking agents can run inside it next to Devin. [59] [78] Stacked pull requests land bottom-up with Devin resolving conflicts between layers, on GitHub.com only. [79] It keeps answering review comments on its PRs until the session is archived. [10]
On deployment, Enterprise buyers get Cognition's Customer Dedicated Deployment, which places the Devbox in a single-tenant VPC reached over PrivateLink or IPSec while the planning brain stays in Cognition's cloud. [53] Outposts, on Pro, Max and Teams, runs sessions on machines you operate with the brain still remote. [80] Jiro's hosted shape is one single-tenant broker for each organisation, with a dedicated deployment available inside your own AWS account. [81] [82] Different shapes; pick the one your security team already understands.
How do you move from Devin to Jiro?
Five things move: repositories, Slack, instructions, the environment, and credentials. Repositories come across through the GitHub App, which you scope to every repo or a chosen handful, or through a stored PAT. [77] Slack needs no migration; you mention a different bot, and every reply in the thread continues that session. [83] Devin's Knowledge, playbooks and Skills become a Nori skillset, authored once and translated into whichever instruction file the agent reads: CLAUDE.md, AGENTS.md or Copilot's. [84] The environment blueprint turns into a setup script. [64] Secrets are re-entered as credentials. Claude subscribers generate their token with claude setup-token before pasting it in. [74] Our switching guide has the full mapping table and a first-week plan, and the head-to-head comparison covers the two products row by row.
Jiro at a glance
| Fact | Jiro |
|---|---|
| What it is | A background coding agent: it works in the cloud and returns pull requests for review [5] |
| Made by | Nori (Tilework Inc.), founded by Amol Kapoor, ex-Google Research [85] |
| Agents it runs | Claude Code, Codex, Cursor, OpenCode, pi-acp [63] |
| Whose model bill | Your Claude/ChatGPT login or API key; Cursor key [1]; Your keys, zero token markup [46] |
| Price | Free 30-day trial; Developer $99/mo; Team $250/mo [2] |
| Runtimes | Developer: up to 3; Team: 5 shared; trial: 5 for 30 days [2] [47] |
| Where work starts | Slack, web, Discord, terminal, schedules, webhooks [4] |
| Where it runs | Isolated cloud environment with your repos and tools [3]; Single-tenant per organization [81] |
| Repos | GitHub App (all or selected repos) or a PAT [77] |
| Not to be confused with | Amazon's Kiro (IDE and agent) or Atlassian's Jira (issue tracker) |
| Last verified | 2026-11-06 |
About the publisher
Jiro is made by Nori (Tilework Inc.), a New York company. Nori was founded by Amol Kapoor, its CEO, who before Nori did AI research at Google on graph neural networks and co-authored MixHop (ICML 2019). [85] [86] Clifford Ressel is Nori's co-founder and CTO. [87] Nori's customers include Corvus and Miravoice. [88] [66]
FAQ
What are the best Devin alternatives?
For a small engineering team whose company already pays Anthropic or OpenAI: Jiro first, since it puts Claude Code, Codex or Cursor Agent on your own credentials without token markup. [46] Then: Cursor's cloud agents if the team lives in Cursor, Codex cloud when it is ChatGPT-only, Anthropic's hosted Claude Code for solo use, Factory as a full platform, Copilot's cloud agent where procurement already has Copilot, Tembo, and OpenHands when you want the source.
How much does Devin cost per month?
There is a Free tier; Pro is $20 monthly and Max is $200 monthly; Teams runs $40 a month per full seat with an $80 monthly minimum; usage past quota is paid with on-demand credits. [7] Enterprise is a custom contract billed in ACUs. [52]
Does Devin still bill in ACUs?
Only on Enterprise. Self-serve plans moved to daily and weekly quotas with dollar-priced credits on 2026-04-14. [52] [48]
Can I use my Claude Max subscription instead of paying Devin for tokens?
Not in Devin: it links personal ChatGPT plans, and that covers the GPT share of usage alone. [58] In Jiro, yes. Generate a token with claude setup-token on your laptop (it lasts a year), then connect it. [74] [1]
Which Devin alternative supports GitLab or Microsoft Teams?
Factory takes delegations from Microsoft Teams, Cursor's cloud agents work with GitLab and Bitbucket repos, and Tembo and OpenHands list GitLab too. [29] [14] [39] [44] Jiro's documented repo connection is GitHub, and Teams is absent from its sourced entry points. [77] [4]
Is Windsurf part of Devin now?
Yes. Cognition introduced Devin Desktop as the next generation of Windsurf on 2026-06-02, and it can host other ACP agents alongside Devin. [59] [78]
Does Jiro use Devin's models or its own?
Neither. Jiro ships no model; it hosts one of five harnesses, and the model is whatever your credential for that harness provides. [63] [1]
Try Jiro
No card needed. Or install Jiro's engineering skills into your own agent: npx nori-skillsets install public/jiro
Sources
- Claude Code in Nori accepts an Anthropic API key, a Claude OAuth login or a setup token; Codex accepts an OpenAI API key or Codex OAuth; Cursor Agent needs a Cursor API key; OpenCode and pi-acp take a compatible API base URL, API key and model. https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
- Nori's plans start with a free 30-day trial with no card; Developer is $99 a month for one user with up to three persistent runtimes, and Team is $250 a month for multiple users sharing five runtimes. https://noriagentic.com/#pricing (checked 2026-11-06)
- Each Nori agent works in an isolated cloud environment with your repository, tools, dependencies and services ready to use. https://noriagentic.com/#faq (checked 2026-11-06)
- Nori work can start from Slack, the web dashboard, Discord, the Handroll terminal, schedules and webhooks, all reaching the same fleet. https://docs.noriagentic.com/start/what-can-nori-do#start-agent-work-from-the-tools-your-team-already-uses (checked 2026-11-06)
- Nori agents can return a pull request, commit, comment or completed task for you to review, merge or send back. https://noriagentic.com/#faq (checked 2026-11-06)
- Devin offers models from Anthropic, OpenAI, Google and Cognition's own SWE family, with an Adaptive option that routes each task to a model automatically. https://docs.devin.ai/cli/models.md (checked 2026-11-06)
- Devin's self-serve plans are Free, Pro at $20 a month, Max at $200 a month and Teams at $40 a month per full seat with an $80 monthly minimum, so one seat and two seats both cost $80 a month; usage past the quota is billed as on-demand credits. https://docs.devin.ai/admin/billing/self-serve.md (checked 2026-11-06)
- Devin Cloud runs sessions on Cognition-hosted machines; Devin CLI and Devin Desktop run locally. https://docs.devin.ai/llms.txt (checked 2026-11-06)
- Devin connects to GitHub, GitLab, Bitbucket and Azure DevOps repos and takes work from Slack, Microsoft Teams, Linear, Jira, PagerDuty and Pylon. https://docs.devin.ai/integrations/overview.md (checked 2026-11-06)
- Devin opens pull requests and automatically responds to comments on PRs its sessions are working on until the session is archived. https://docs.devin.ai/integrations/gh.md (checked 2026-11-06)
- Cloud agents use your plan's usage like other agent work, at the selected model's rate. https://cursor.com/docs/cloud-agent (checked 2026-11-06)
- Cursor subscriptions are sold only through X Subscription Corp; Premium at $30 a month is the first tier that includes cloud agents and automations, and third-party models are billed separately at API list price. https://cursor.com/pricing (checked 2026-11-06)
- Cursor cloud agents run in isolated cloud VMs with full development environments, or move tool execution to self-hosted machines you manage. https://cursor.com/docs/cloud-agent (checked 2026-11-06)
- Cloud agents start from the Cursor desktop app, the web, iOS and Android, Slack, GitHub and Bitbucket comments, Linear and an API, on GitHub, GitLab, Bitbucket Cloud or Azure DevOps repos. https://cursor.com/docs/cloud-agent (checked 2026-11-06)
- Cursor cloud agents produce merge-ready PRs with screenshots, videos and logs, and open a PR in each repo they touch. https://cursor.com/docs/cloud-agent (checked 2026-11-06)
- Codex cloud runs OpenAI models; an API key unlocks Codex in the CLI, SDK and IDE but not cloud features, which need a ChatGPT plan. https://learn.chatgpt.com/docs/pricing.md (checked 2026-11-06)
- Codex cloud is included from ChatGPT Plus at $20 a month; Pro costs $100, $200 or $500 a month and Business $25 per user a month billed monthly or $20 billed annually, with usage limits shared with ChatGPT. https://learn.chatgpt.com/docs/pricing.md (checked 2026-11-06)
- Codex cloud runs each task in a cloud VM, so work continues while your computer is asleep. https://learn.chatgpt.com/docs/environments/cloud-environments.md (checked 2026-11-06)
- Codex cloud tasks start from ChatGPT on the web, mobile and desktop, or from the Codex CLI. https://learn.chatgpt.com/docs/cloud.md (checked 2026-11-06)
- In Codex cloud you review the changes, then commit or open a pull request when you are ready. https://learn.chatgpt.com/docs/cloud.md (checked 2026-11-06)
- Cloud sessions run Claude models and require an Anthropic account; they are not available through Bedrock, Google Cloud or other third-party providers. https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06)
- Claude Code is included in Claude Pro ($20 a month, or $17 a month billed annually), Max ($100 a month), Ultra ($200 a month) and Team and Enterprise seats; the Free plan has no Claude Code. https://claude.com/pricing (checked 2026-11-06)
- A Claude Code cloud session runs in a fresh Anthropic-managed VM with 4 vCPUs, 16 GB of RAM and 30 GB of disk, or in an organization's self-hosted environment when routed there. https://code.claude.com/docs/en/cloud-environments (checked 2026-11-06)
- Cloud sessions start from claude.ai/code, the Claude mobile and desktop apps, and
claude --cloudin a terminal, andclaude --teleportpulls a session back to your machine. https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06) - Cloud sessions push a branch; you review the diff and create the pull request, and auto-fix can then push fixes for CI failures and review comments. https://code.claude.com/docs/en/claude-code-on-the-web (checked 2026-11-06)
- Factory routes across Anthropic, OpenAI and other models and accepts your own keys or local models in the CLI and desktop app, but not in the hosted web or mobile apps. https://docs.factory.com/model-independence/byok.md (checked 2026-11-06)
- Factory's plans are Pro at $20 a month, Plus at $100, Max at $200, and Teams at $60 a month plus $40 per seat for up to 10 seats; Business and Enterprise are custom. https://factory.com/pricing (checked 2026-11-06)
- Droids run locally in the Droid CLI and Factory App, on Factory-hosted Droid Computers, on your own computer via Remote Control, or in your own cloud via OpenCompute. https://docs.factory.com/llms.txt (checked 2026-11-06)
- Droids work in the Factory App, Droid CLI, headless Droid Exec and IDEs, and take delegations from Slack, Linear, Jira and Microsoft Teams. https://docs.factory.com/llms.txt (checked 2026-11-06)
- When delegated a Linear issue, a Droid is instructed to open a pull request and return its URL. https://docs.factory.com/delegations/linear.md (checked 2026-11-06)
- The cloud agent offers Auto plus GitHub-provided Claude, Gemini, GPT and MAI models; bring-your-own-key is documented for IDEs, Copilot CLI, the Copilot app and the SDK, not for the cloud agent. https://docs.github.com/en/copilot/reference/ai-models/supported-models (checked 2026-11-06)
- Copilot Pro is $10, Pro+ $39 and Max $100 a month; Business is $19 and Enterprise $39 a month per user; Copilot Free does not include the cloud agent. https://docs.github.com/en/copilot/get-started/plans (checked 2026-11-06)
- Copilot cloud agent sessions run in a GitHub cloud sandbox, in a GitHub Actions environment capped at 59 minutes, or in a remote environment you connect. https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-cloud-agent (checked 2026-11-06)
- Copilot cloud agent sessions start from GitHub.com, GitHub Mobile, VS Code, JetBrains, Visual Studio and Eclipse, the GitHub CLI, REST API and MCP server, Raycast, Slack, Microsoft Teams, Linear, Jira and Azure Boards. https://docs.github.com/en/copilot/how-tos/use-copilot-agents/cloud-agent/create-a-pr (checked 2026-11-06)
- Copilot cloud agent opens draft pull requests on its own branch and cannot approve or merge them. https://docs.github.com/en/copilot/concepts/agents/cloud-agent/risks-and-mitigations (checked 2026-11-06)
- Tembo runs Claude Code, Codex, OpenCode or Pi per task and accepts your own OpenAI, Anthropic or xAI keys or a Claude, ChatGPT or SuperGrok subscription. https://docs.tembo.io/features/models.md (checked 2026-11-06)
- Tembo's Pro plan is $60 a month with $90 of included usage and Max is $200 a month with $400, with cloud VMs metered by the second from $0.1066 an hour. https://www.tembo.io/pricing (checked 2026-11-06)
- Tembo runs coding agents in isolated Docker or VM sessions with code, terminal, files and a live desktop. https://www.tembo.io/ (checked 2026-11-06)
- Tembo starts work from GitHub, GitLab, Bitbucket, Slack, Linear, Jira, Sentry, PagerDuty, webhooks and schedules. https://www.tembo.io/ (checked 2026-11-06)
- Tembo turns issues and alerts from Sentry, Linear and Jira into pull requests, including coordinated multi-repo PRs across GitHub, GitLab and Bitbucket, and reviewer feedback can be sent back into the same session. https://www.tembo.io/ (checked 2026-11-06)
- OpenHands runs its own agent or Claude Code, Codex, Gemini, Pi, OpenCode or any ACP agent, and works with any LLM. https://github.com/OpenHands/OpenHands (checked 2026-11-06)
- Agent Canvas is free to self-host for a single user; Teams costs $500 a month for up to 10 users with extra seats at $50 a month each and a cap of 5,000 sessions a month, and Enterprise is custom. https://openhands.dev/pricing (checked 2026-11-06)
- OpenHands runs locally by default, with agents running locally, in Docker, on VMs, or optionally on OpenHands Cloud or OpenHands Enterprise infrastructure. https://github.com/OpenHands/OpenHands (checked 2026-11-06)
- OpenHands Cloud works from its web UI and API, from GitHub, GitLab and Bitbucket repos, and from Slack. https://docs.openhands.dev/openhands/usage/cloud/cloud-ui (checked 2026-11-06)
- Labelling a GitHub issue openhands or commenting @openhands starts a run that opens a PR if it resolves the issue. https://docs.openhands.dev/openhands/usage/cloud/github-installation (checked 2026-11-06)
- Nori's comparison table lists any model on your own keys with flat pricing and zero token markup. https://noriagentic.com/#compare (checked 2026-11-06)
- Nori sizes plans by runtimes, which sleep when idle and wake on demand; the trial runs five for 30 days. https://noriagentic.com/#faq (checked 2026-11-06)
- Devin Pro and Teams full seats get a daily and a weekly usage quota, Max gets a larger weekly quota with no daily cap, the Free plan gets limited usage plus Devin Review and DeepWiki, and usage beyond quota is paid with on-demand credits that never expire. https://docs.devin.ai/admin/billing/self-serve.md (checked 2026-11-06)
- Free and Pro allow up to 10 concurrent Devin sessions; Max and Teams have no concurrency cap. https://devin.ai/pricing (checked 2026-11-06)
- A Devin Teams account holds up to 200 members; full seats cost $40 a month each with a Pro-sized quota, flex seats are free and draw on a shared on-demand credit pool, and when fewer than two full seats are paid the rest of the $80 minimum is charged as prepaid credits. https://docs.devin.ai/admin/billing/self-serve.md (checked 2026-11-06)
- On Devin's self-serve plans the Adaptive router draws down quota at a fixed $0.50 per million input tokens and $2.00 per million output tokens whichever model it picks, and enterprise accounts have Adaptive switched off until an admin enables it. https://docs.devin.ai/cli/adaptive.md (checked 2026-11-06)
- Cognition retired its Core and Team plans for quota-based self-serve plans on 2026-04-14, billing self-serve overage in dollars instead of ACUs while Enterprise stays on ACUs. https://cognition.com/blog/new-self-serve-plans-for-devin (checked 2026-11-06)
- In Devin's Customer Dedicated Deployment, Cognition hosts the Devbox in a customer-isolated single-tenant VPC reached from your network over AWS PrivateLink or an IPSec tunnel, and the Devin brain always resides in Cognition's cloud. https://docs.devin.ai/enterprise/deployment/overview.md (checked 2026-11-06)
- Usage past a Pro, Max or Ultra plan's limits can continue as usage credits billed at standard API rates. https://support.claude.com/en/articles/12429409-extra-usage-for-paid-claude-plans (checked 2026-11-06)
- Claude Max costs $100 a month for five times the Pro plan's usage every five hours and every week; the former Max 20x tier is now sold as Ultra. https://support.claude.com/en/articles/11049741-what-is-the-max-plan (checked 2026-11-06)
- Since 2026-06-01, all Copilot plans bill in GitHub AI Credits, which meter input, output and cached tokens at per-model rates, with one credit equal to $0.01. https://docs.github.com/en/copilot/concepts/billing-and-usage/individuals/billing (checked 2026-11-06)
- Tembo Pro covers up to 5 users and Max up to 10, both with overage billing past the included usage; the only free option is a 7-day trial of Max, and Enterprise is custom. https://www.tembo.io/pricing (checked 2026-11-06)
- On Devin's self-serve Pro, Max and Teams plans a user can link a personal ChatGPT Go, Plus or Pro plan so that requests to OpenAI models bill to that plan, while non-GPT models and cloud machine time still consume Devin quota. https://docs.devin.ai/admin/billing/chatgpt.md (checked 2026-11-06)
- Cognition introduced Devin Desktop as the next generation of Windsurf on 2026-06-02. https://cognition.com/blog/introducing-devin-desktop (checked 2026-11-06)
- Devin Knowledge is a collection of instructions Devin can reference in all sessions, pinned to repositories or retrieved by a trigger description; Cognition has deprecated it and is migrating existing entries to Skills in Plugins. https://docs.devin.ai/onboard-devin/knowledge-onboarding.md (checked 2026-11-06)
- Devin playbooks are reusable prompts for repeated tasks, attached to a session by typing a !-prefixed macro, and Cognition is replacing them with Skills, which new accounts start with from 2026-10-24. https://docs.devin.ai/product-guides/creating-playbooks.md (checked 2026-11-06)
- Nori cron triggers run recurring jobs, reminders run once at a future time, and webhooks react to external events. https://docs.noriagentic.com/resources/faq#can-nori-run-tasks-without-a-person-online (checked 2026-11-06)
- Nori's default first-class Sessions agents are Claude Code, Codex, Cursor Agent, OpenCode and pi-acp. https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
- A Nori organization can define a setup script that installs dependencies and prepares every session's environment. https://docs.noriagentic.com/start/what-can-nori-do#give-agents-a-real-working-environment (checked 2026-11-06)
- In Nori, one claimed machine serves one active session within the organization's fleet. https://docs.noriagentic.com/resources/faq#is-each-session-isolated (checked 2026-11-06)
- Nori handles more than 25% of Miravoice's merged pull requests, and Miravoice's CTO says it easily saves him five hours a week. https://noriagentic.com/newsletter/2026-09-18-miravoice-case-study.html (checked 2026-11-06)
- Cursor's help page on background agents says that in Cursor these are called Cloud Agents. https://cursor.com/help/ai-features/background-agents (checked 2026-11-06)
- Codex requests in Slack now start with @ChatGPT instead of @Codex. https://learn.chatgpt.com/docs/third-party/slack.md (checked 2026-11-06)
- For Team and Enterprise, Slack work moves to Claude Tag (public beta), and the legacy Claude Code Slack bot was retired on 2026-10-05 in workspaces connected to a Claude organization. https://code.claude.com/docs/en/slack (checked 2026-11-06)
- Factory automation starters such as Ticket to PR, Slack Bug Fixer and Alert Responder run on schedules or events and open pull requests. https://docs.factory.com/software-factory/automations.md (checked 2026-11-06)
- Factory Private, announced 2026-09-18 and still in private preview, places some or all of Factory's infrastructure in the customer's VPC. https://factory.com/news/factory-private (checked 2026-11-06)
- GitHub's Copilot cloud agent, now presented as Copilot in the cloud, researches a repository, plans changes and implements them in a GitHub cloud sandbox, a GitHub Actions environment or a remote environment you connect. https://docs.github.com/en/copilot/concepts/agents/cloud-agent/about-cloud-agent (checked 2026-11-06)
- OpenHands is an MIT-licensed open-source, self-hosted control center for coding agents and automations. https://github.com/OpenHands/OpenHands (checked 2026-11-06)
- The claude setup-token command generates a one-year token tied to a Pro, Max, Ultra, Team or Enterprise plan for machines where browser login isn't available. https://code.claude.com/docs/en/authentication (checked 2026-11-06)
- Anthropic reports that across enterprise deployments Claude Code averages about $13 per developer per active day and $150 to $250 per developer per month, staying below $30 per active day for 90% of users. https://code.claude.com/docs/en/costs (checked 2026-11-06)
- Devin Automations start sessions from schedules, webhooks, and Slack, GitHub, GitLab, Linear, Jira, Pylon and PagerDuty events. https://docs.devin.ai/product-guides/automations.md (checked 2026-11-06)
- Nori supports two GitHub credential methods: a GitHub App you install and scope to all repositories or a selected set, or a saved personal access token. https://docs.noriagentic.com/sessions/guides/integrations#github (checked 2026-11-06)
- Devin Cloud runs Cognition's Devin agent; Devin Desktop can also host other agents that speak ACP. https://cognition.com/blog/introducing-devin-desktop (checked 2026-11-06)
- Devin splits large changes into stacked pull requests that land bottom-up, resolves conflicts between layers itself and fixes CI across the stack, and stacking works only on GitHub.com repositories. https://docs.devin.ai/work-with-devin/stacked-prs.md (checked 2026-11-06)
- Devin Outposts, available on Pro, Max and Teams, runs sessions on machines you operate through a self-hosted worker that needs only outbound HTTPS, while planning and inference stay in Devin's cloud. https://docs.devin.ai/cloud/outposts/overview.md (checked 2026-11-06)
- Each hosted Nori organization gets a single-tenant broker on its own subdomain. https://docs.noriagentic.com/sessions/concepts/security-and-credentials#organization-isolation (checked 2026-11-06)
- Nori hosts Jiro in its own cloud, and a dedicated BYOC deployment runs private EC2 session machines inside customer-controlled AWS infrastructure. https://docs.noriagentic.com/sessions/reference/supported-agents-and-compute (checked 2026-11-06)
- Mention Nori's bot in a Slack channel to start a session; it replies in a thread, and every reply in that thread continues the same session. https://docs.noriagentic.com/sessions/guides/slack (checked 2026-11-06)
- A Nori skillset is written once and translated into each agent's own files: CLAUDE.md for Claude Code, GEMINI.md for Gemini CLI, copilot-instructions.md for GitHub Copilot and AGENTS.md for the rest. https://docs.noriagentic.com/skillsets/concepts/translation-and-agents (checked 2026-11-06)
- Nori is led by CEO Amol Kapoor, a former Google Research scientist, and CTO and co-founder Clifford Ressel. https://ai.engineer/speakers/amol-kapoor (checked 2026-11-06)
- Amol Kapoor's Google-affiliated research includes MixHop (ICML 2019), N-GCN and a 2020 spatio-temporal graph neural network for COVID-19 forecasting. https://scholar.google.com/citations?user=SC_-qB8AAAAJ (checked 2026-11-06)
- Clifford Ressel is Nori's CTO and co-founder. https://ai.engineer/orgs/nori (checked 2026-11-06)
- Corvus's CTO saw a production bug at a client dinner, had Nori look into it and deploy a fix. https://noriagentic.com/newsletter/2026-05-19-corvus-case-study.html (checked 2026-11-06)